See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/257390915

Synthesis and characterization of SnO2 thin films prepared by dip-coating

method

Article  in  Physics Procedia · December 2012

DOI: 10.1016/j.phpro.2012.03.664

CITATIONS READS

23 1,374

9 authors, including:

Marco Antonio Schiavon M. T. Raposo

Federal University of São João del-Rei Federal University of São João del-Rei
139 PUBLICATIONS   2,175 CITATIONS    15 PUBLICATIONS   56 CITATIONS   

SEE PROFILE SEE PROFILE

R. de Paiva Jose Luiz Aarestrup Alves

Federal University of São João del-Rei Federal University of Minas Gerais
23 PUBLICATIONS   388 CITATIONS    61 PUBLICATIONS   693 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Catalytic oxidation of aqueous sulfide View project

Nanobiotecnology Project and Innovaton View project

All content following this page was uploaded by José D Ardisson on 31 May 2014.

The user has requested enhancement of the downloaded file.

 Available online at www.sciencedirect.com

Physics Procedia 28 (2012) 22 – 27

15th Brazilian Workshop on Semiconductor Physics (BWSP-15)

Synthesis and characterization of SnO2 thin films prepared by dip-coating

method

D. H. Q. Carvalhoa*, M. A. Schiavona, M. T. Raposoa, R. de Paivaa, J. L. A. Alvesa, Roberto. M. Paniagob,

N. L. Spezialib, A. S. Ferlautob, J. D. Ardissonc

a
Departamento de Ciências Naturais - Universidade Federal de São João del-Rei, MG, Brazil
b
Departamento de Física - Instituto de Ciências Exatas - Universidade Federal de Minas Gerais
c
Laboratório de Física Aplicada - Comissão Nacional de Energia Nuclear - Centro de Desenvolvimento da Tecnologia Nuclear.

Abstract

The optical, electrical and structural properties of SnO2 are responsible for a large number of technological
applications such as gas sensors, optical-electonic devices, varistors and displays. In this paper, we report the
preparation of SnO2 thin films deposited on glass, quartz and silicon substrates by the technique of sol-gel dip-
coating. The films were characterized by X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS),
ellipsometry and Mössbauer spectroscopy. We combine the experimental results with ab initio all-electrons
calculations , using the density functional theory within the framework of the full-potential linear augmented
plane waves method, in order to extract hyperfine parameters. The results show that the synthesis method is able to
produce good quality films and that the theory can be helpful to determine quantities difficult to be measured
experimentally.
© 2012 Published by Elsevier B.V. Selection and/or peer-review under responsibility of Universidade Federal de
© 2011
Juiz Published
de Fora, by Elsevier B.V.
Brazil.
Pacs:

Keywords: SnO2, synthesis, characterization, thin films, dip-coating

Introduction

SnO2 films , due to their good transparency to light ( fundamental energy gap ~3.6 eV) and high chemical
stability have many applications in the manufacturing of transparent electrodes , flat panel displays [1], coatings
for heat reflection [2], solar cells [2] and gas sensors [3] . These applications require conductive layers with optical
and electrical properties suited to specific applications. On the other hand, the properties of SnO2 films are strongly
dependent on the crystallographic quality and surface morphology of the deposited layers [4]. Different techniques
have been employed to obtain pure and doped layers of SnO2, such as chemical vapor deposition [5], aerosol
pyrolysis [6], sputtering [7], physical vapor deposition [8], laser abrasion [9], and chemical routes by using sol-gel
[10]. The latter approach represents a low temperature and low-cost one and allows fine control of composition,
microstructure, and morphology of the obtained layers [11]. Nanostructured films of pure and doped SnO2 have
been obtained from tetra (tert-butoxy) tin (IV), SnCl2 or SnCl4 as precursors and by means of dip-coating [12] and
spin-coating [13]. In this work we choose the sol-gel dip-coating technique following chemical recipes found in
the literature[14,15]; the films were deposited on glass, quartz and silicon substrates and characterized by X-ray
diffraction (XRD ), X-ray photoelectron spectroscopy (XPS), ellipsometry and Mössbauer spectroscopy . We
combine experimental results with ab initio all-electrons calculations , using the Density Functional Theory
(DFT) within the framework of the full-potential linear augmented plane waves (FPLAPW) method, in order to

1875-3892 © 2012 Published by Elsevier B.V. Selection and/or peer-review under responsibility of Universidade Federal de Juiz de Fora, Brazil.
doi:10.1016/j.phpro.2012.03.664
 D. H. Q. Carvalho et al. / Physics Procedia 28 (2012) 22 – 27 23

extract hyperfine parameters. This is a preliminary experimental and theoretical work aiming the study of the
influences of defects and dopants on the performance of such films as gas sensors.

Methodology

Experimental

The pure SnO2 films were prepared by the technique of sol-gel dip-coating by following physical [14] and
chemical procedures existing in the literature [15]. The method of dip-coating [16] is adequate for the preparation of
films having good structural qualities and has the advantages of using high purity materials and being easy to
produce coatings at low cost on a variety of substrates . The technique consists of dipping a substrate onto a
solution, a sol-gel containing the precursors, and then of removing it at a controlled speed. The SnO2 films were
grown on glass, quartz, and silicon substrates and were characterized by X-ray diffraction (XRD), X-ray
photoelectron spectroscopy (XPS), and ellipsometry. 119Sn Mössbauer spectroscopy was used to provide
morphological information of the films, the local neighborhood and oxidation states of the Sn atoms, the isomer
shifts and quadrupole splittings.

Theoretical

We used the Density Functional Theory [17] as implemented in the WIEN2K -FPLAPW computational code [18],
which takes into account all the electrons of the system, as opposed to the pseudo-potential approaches, in order
to carry out the calculations of the atomic and electronic structures of bulk SnO2 in its tetragonal rutile phase. From
these calculations we obtained the isomer shift, į , the electric field gradient , Vzz , and the asymmetry parameter,
Ș, to be associated to the experimental quadrupole splittings ,ǻ , in order to evaluate the quadrupole moment , Q,
of the Sn nucleus through ǻ=( ½)eŇQVzzŇ(1+1/3 Ș)1/2 , where e is the elementary charge .
The optimized lattice parameters, obtained using the Generalized Gradient Approximation (GGA), were a = b =
4.71 (Å), c = 3.23 (Å) and u = 0.307. The input paremeters for the WIEN2K code were taken Rkmax = 8.0, Lmax =
10, Kgen = 500, fc = 1 (mRy/au) and ec = 0.0001 (Ry)

Results and Discussion

X-ray photoelectron spectroscopy (XPS)

The XPS measurement is a very sensitive technique that can tell us clearly the quality of the surface of deposited
thin films. Our XPS measurements were performed using a VG ESCALAB 220-iXL photoelectron spectrometer
with a dual anode X-ray source (Al K-alpha/1486.6eV, Mg K-alpha/1253.4), and a spherical electron analyzer
equipped with 5 channeltrons. The XPS spectra were collected with aid of the Avantage software from Thermo
Scientific. A spectral analysis of each sample was performed in the 0-1000 eV range (survey spectrum) to identify
the chemical elements on the surface. Detailed spectra of O 1s and Sn 3d states were collected to determine the
relative concentration of the elements present in each sample as well as to identify their binding energies with high
energy resolution (r0.1eV). The C1s state was also recorded for each sample in order to evaluate surface charging
effects during spectra acquisition and to correct the binding energy scale. Given the kinetic energy of the
photoelectrons, the estimated probed depth of the samples was within 5-9 monolayers. Spectroscopic analyses were
performed using Gaussian-Lorentzian peak fitting, and by applying the Shirley method to subtract background. We
have studied five samples and the analyses showed that besides oxygen and tin, the films also contained very small
quantities of chlorine. In Fig.1 we show the XPS spectrum for a sample of a SnO2 film deposited on a quartz
substrate.
24 D. H. Q. Carvalho et al. / Physics Procedia 28 (2012) 22 – 27

Independently of the film thickness and of the

particular substrate used, we observed a low amount
of chlorine in all of them. The appearance of Cl is
probably related to the use of the 2SnCl2. 2H2O
precursor. Cl can be removed through the dialysis
procedure and heat treatment. The presence of carbon
is probably due to organic contaminants in the films
after deposition. The detailed analysis of the C1s
peak for all samples revealed a shift of ~1.0 eV to
higher binding energies. Assuming that this shift is
due to surface charging, the corrected binding energy
of the Sn-3d5/2 peak is at 487.4 eV (see Fig. 2), which
is very close to the value reported for the binding
energy of Sn 3d5/2 for SnO2, found in the literature,
which is 486.6 eV [19]. The asymmetric shape of the
O1s-peak (see Fig. 3) is explained by the existence of
Figure1. XPS spectrum of SnO2 film grown on quartz substrate.
other oxygen species at the surface (adsorbed CO or
CO2 and organic contaminants) which add a shoulder
at the left side (higher binding energy side) of the
O1s peak of SnO2.

Figure 3. XPS spectrum of SnO2 film grown on quartz

Figure 2. XPS spectrum of SnO2 film grown on quartz substrate. substrate. The peak of the 1s-O state is shown.
The peaks of the Sn3/2 and Sn5/2 states are shown.

When the samples were subjected to a bombardment of argon ions in order to remove surface contamination,
the oxygen 1s peak approaches a symmetrical shape, suggesting that oxygen in different oxidation states is
responsible for the weak asymmetry ( the electrons corresponding to the removed species having a higher binding
energy than the energy of the 1s-O state in SnO2) ) . Otherwise, these types of oxygen may correspond to organic
contaminants adsorbed on the surface, since carbon was also found on the samples. Actually, the spectra exhibited a
remarkable shift to higher energies as an effect of loading when getting the spectra ; the main peak 1s-O (SnO2) ,
expected to be around 530.6eV, shows up at 531.6 eV, the loading effect being about 1.0 eV, as found for 1s-C
state.

X-ray Diffraction

The X-ray diffraction (XRD) technique was used in order to study the crystalline structure of the thin films and
identify the preferred orientations of crystallites of SnO2. The diffraction data of the as-deposited thin films were
collected using a RIGAKU-Geigerflex diffractometer, in ș-2ș Bragg-Brentano geometry. The measurements were
made by focusing the radiation at a fixed grazing incidence angle ș = 1.0o. The diffraction patterns are shown in
Figs.4 to 6 and all of them exhibit , besides the typical peaks of the substrates , the main characteristic peaks
 D. H. Q. Carvalho et al. / Physics Procedia 28 (2012) 22 – 27 25

(110), ( 101), and (211) of the SnO2 rutile phase ; the peaks have been identified by comparison with JCPDS data
(Fig.7).

Figure 4.XRD pattern of SnO2 film deposited on glass substrate. The Figure 5. XRD pattern of SnO2 film deposited on quartz substrate.
(*) symbol refers to the substrate’s peaks. The (*) symbol refers to the substrate’s peaks.

Figure 6. XRD s pattern of SnO2 film deposited on silicon Figure 7. Standard XRD spectrum of rutile phase of SnO2
substrate. The (*) symbol refers to the substrate’s peaks.

Based on the measured diffraction patterns and on the JCPDS intensities we calculated the texture coefficients
(TC) [20] for each sample which are displayed in Table I. A deviation of the texture coefficient from unit implies
the preferred orientation of the growth; the larger the texture coefficient deviation from unit, the higher is the
preferred orientation of a film. As can be observed from Table I, all samples exhibit a (211) preferred orientation; it
is apparent the little influence of the substrates upon the texture coefficients. The coefficients are close to unit, what
means a weak texture of the films. We have also used the Scherrer´s equation in order to estimate the average sizes
of the crystallites of SnO2 in the thin films; in Table I we also display these estimates. It is interesting to note that the
smallest crystallites have the (211) orientation at large.

TableI. SnO2 thin films : texture coefficients ,TC, crystallite sizes(Å), and plane spacings , d(h,k,l)(Å).

*JCPDS, “Joint Comittee on Powder Diffraction Standards”.

Ellipsometry analysis

Ellipsometry is an optical non-destructive technique allowing the accurate characterization of thin films by
measuring the change of light polarization upon reflection on a sample. We used this technique in order to evaluate
the thicknesses of the as-grown SnO2 films with 8 dips on the glass, quartz and silicon substrates. Ellipsometry
measures the complex reflectance ratio, ȡ = tang(ȥ)exp(iǻ) which may be parametrized by Ȍ and ǻ. A layer model
of the sample must be established, which considers the optical constants and thickness parameters of all individual
26 D. H. Q. Carvalho et al. / Physics Procedia 28 (2012) 22 – 27

layers. Using an iterative procedure, unknown optical constants and/or thickness parameters are varied, and Ȍ and ǻ
values are calculated using the Fresnel equation the calculated Ȍ and ǻ values which match the experimental data
lead to the optical constants and thickness parameters of the sample. An equipment Wooller , model M2000 ,
having a wavelength range 245-1690 nm, was used to make measurements at three angles The corresponding
ellipsometric parameters , Ȍ and ǻ , were determined, thus leading to the values of the film thicknesses . The
substrates were modeled using uncoated samples and the thickness was determined by linear regression analysis.
The experimental Ȍ and ǻ data and the corresponding model fits are shown in Fig.8.

Figure 8. Experimental ellipsometric data for (up) and ( down) vs. wavelength, with the model fits (dotted lines), taken at three different angles
(colors), for SnO2 films deposited on quartz substrate after 8 dips.

It is worth mentioning that the model does not provide a perfect fit for the experimental data, probably because of
the use made of bulk optical constants to model the SnO2 film. The fit of the Ȍ and ǻ curves as a function of
wavelength were carried out by means of three Lorentzian curves. Once the adjustments were made, a comparison
between the experimental curves (solid lines) and the theoretical model ones (dotted lines) provided a theoretical
best fit and the corresponding stored theoretical data permitted to obtain the value of the thickness of each film .
The SnO2 film thicknesses determined by means of this kind of analysis were 123.59 nm, 156.89nm, and 97.88nm
for glass, quartz, and silicon substrates, respectively.

Mössbauer spectroscopy

All of the 119Sn Mössbauer results for the SnO2 films were recorded using Conversion Electron Mössbauer
Spectroscopy (CEMS) at room temperature and using CaSnO3 as source. Transmission mode Mössbauer
spectroscopy is prohibited in our cases due to the attenuation of the Mössbauer gamma ray by the substrates. The
spectra were recorded with the sample plane normal to the incident gamma ray direction and are shown in Fig. 9.
The hyperfine parameters from least squares fits are given in Table II.

Figure 9. CEMS spectra for samples prepared on glass (A01), quartz (A02) and silicon (A03) with 8 dips and on quartz (A04), and silicon (A05)
with 4 dips.
 D. H. Q. Carvalho et al. / Physics Procedia 28 (2012) 22 – 27 27

Table II.Hyperfine parameters, isomer shift, į, quadrupole

splitting, ', line width, *, and quadrupole moment Q.

Figure 10. CEMS spectra for samples prepared on quartz (A02)

and silicon (A03), with 8 dips, and the corresponding quadrupole
splitting distributions P (ǻ).

As anticipated by the literature, the isomer shifts and quadrupole splits are close to the values for the CaSnO3
source [21]. The distribution of the quadrupole splits, Fig. 10, exhibit a distortion of the electron density around the
nucleus , probably associated to the presence of tin atoms in states other than the Sn+4 in the lattice of the
SnO2 crystal . Also shown in Table II are the quadrupole moments, Q, as obtained by combining the experimental ǻ
values and the theoretical electric field gradient, Vzz, calculated for the bulk. These values for the several samples
are very close to the experimental value ŇQŇ = 10.9 ± 0.8 fm2 [22].

Final remarks

The stoichiometry of tin dioxide was verified for all samples by XPS which also revealed very small quantities of
chlorine and carbon as contaminants.The XDR diagrams recorded under grazing incidence showed that the layers
are polycrystaline and that the peak intensities correspond fairly well to the intensities given by the JCPDS card.
This is indicative of a random crystallite orientation (i.e. powder-like sample), though the (110) orientation be
preferred. After eight dips, the obtained films were ~ 100 nm thick, as measured by ellipsometry. The Mössbauer
spectra showed that SnO2 is present in the films as single phase, the isomer shift is close to that of CaSnO3, as
expected, and the experimental/theoretical quadrupole moment obtained for the tin nucleus is close to the
experimental value reported in the literature.
References
1. Tsirlina GA, Roginskaya YuE, Postovalova GG, Vassiliev SYu Russ J Electrochemistry 35, 1385. (1999).
2. Courtney IA, Dahl JR. J Electrochem Soc 144, 2045(1997); 144, 2943 (1997).
3.Zhu J, Lu Zh, Aruna ST, Aurbach D, Gedanken A Chem Matter 12,2557 (2000) .
4. A. Nuruddin and J.R. Abelson, Thin Solid Films 394, 49 (2001).
5. Barreca, D.; Garon, S.; Zanella, P.; Tondello, E. J. Phys. IV France, 9, 667(1999).
6. Matko, I.; Gaidi, M.; Hazemann, J. L.; Chenevier, B.; Labeau, M. Sensors Actuators
7. Gaggiotti, G.; Galdikas, A.; Kaiulis S.; Mattogno, G.; Setkus, A. Sensors Actuators B, 24í25, 516í519 (1995).
8. Schweizer-Berberich, M.; Zheng, J. G.; Weimar, U.; Göpel, W.; Bârsan, N.; Pentia, E.; Tomescu, A. Sensors Actuators B, 31, 71í75 (1996).
9. Williams, G.; Coles, G. S. V. MRS Bull., 24 (6), 25í29 (1999).
10. Brinker, C. J.; Scherer, G. W. In SolíGel Science: The Physics and Chemistry of SolíGelProcessing; Academic Press: New York, 1990.
11 Varghese, O. K.; Mahlotra, L. K.; Sharma, G. L. Sensors Actuators B, 55, 161í165(1999).
12.M. Acciarri, C. Canevali, C. M. Mari, M. Mattoni, R. Ruffo, R. Scotti, and F. Morazzoni,D. Barreca, L. Armelao, and E. Tondello , E.
Bontempi and L. E. Depero .Chem. Mater, 15 , 2646 (2003).
13. Shimizu, Y.; Egashira, M. MRS Bull., 24, 18 (1999).
14. Alfredo R. M. Oliveira, Aldo J. G. Zarbin. Quim.Nova , 28,1(2005).
15J. Grim,D.Bessarov, W. Maier, S. Storck and R.D.Saderson. Desalination 115,295(1998).
16. N.J.Arfsten, J.Non. Cryst. Solids, 63,243(1984).
17. Hohenberg,P. and Kohn, W., Phys.Rev.136,B864(1964),; Kohn,W and Sham,L.J., Phys.Rev. 140,A1133(1965).
18.Blaha,P.,Schwarz,K., Sorantin,P., and Trickey,S.B.,Comput.Phys.Commun.59,399(1990),.
19. N. Ikeo. Handbook of X-ray photoelectron spectroscopy. JEOL, 1991.
20.Moholkar, A. V.; Pawar, S. M.; Rajpure, K. Y.; Patil, P. S.; Bhosale J. of Phys. and Chem. of Solids, 68, 1981, (2007).
21. M.Methfessel, C.O.Rodriguez, O.K.Anderson.Phys.Rev.B40, 2009(1989).
22. H.Haas, M.Menningen, H.Andreasen,S. Damgaard, H. Grann,F.T. Pedersen,J.W. Petersen , andG.Weyer, Hyp.Interact. 15/16(1983).

View publication stats