pss a

solidi
phys. stat. sol. (a) 205, No. 8, 2075 – 2079 (2008) / DOI 10.1002/pssa.200778939

physica

status
www.pss-a.com
applications and materials science
ZnO/SiO2 nanocomposite thin films
by sol–gel method
V. Musat*, 1, E. Fortunato**, 2, S. Petrescu1, and A. M. Botelho do Rego3
1
Department of Metals and Materials Science, “Dunarea de Jos” University of Galati, 47 Domneasca, 800008 Galati, Romania
2
Department of Materials Science, CENIMAT, Faculty of Sciences and Technology, New University of Lisbon, Campus da Caparica,
2829-516 Caparica, Portugal
3
Department of Chemical and Biological Engineering, CQFM, Technical University of Lisbon, 1049-001 Lisbon, Portugal

Received 6 August 2007, revised 19 June 2008, accepted 20 June 2008

Published online 21 July 2008

PACS 73.63.Bd, 78.66.Sq, 81.05.Dz, 81.20.Fw

*
Corresponding author: e-mail viorica.musat@ugal.ro, Fax: +40 2 36 460754
**
e-mail elvira.fortunato@fct.unl.pt, Fax: +351 212 948 558

In recent years ZnO/SiO2 nanocomposite thin films have at-
tracted much attention for a wide range of device applications
based on their specific optical and electrical properties.
Mesoporous silica nanocomposite with controllable variation
of size and shape of pores is the most challenging system for
the generation of stable oxide semiconductive nanoparticles
with controlled size and shape and subsequently with con-
trolled optical and electrical properties. This paper presents
the results on mesoprous ZnO/SiO2 nanocomposite thin films
obtained by sol – gel method using three different ways of the
sol preparation. The effect of the sol preparation on the zinc-
doping states of the thin films is discussed based on X-ray
photoelectron spectroscopy. The crystalline structure, mor-
phology, optical transmittance, optical band gap, electrical
properties and room temperature photoreduction-ozone re-
oxidation properties of these films are presented.

© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

1 Introduction Nanostructured ZnO-based semicon-
ductive thin films have a very wade range of applications
based on very interesting properties depending on crystal-
lite size. In recent years, these low-cost and no-toxic thin
films are intensively investigated to replace very expensive
materials in different application [1–7], such as indium
thin oxide (ITO) in solar cells and flat DCL devices [1, 2,
6] and GaN in broadband UV photodetectors with high
tunable wavelength [3]. Important applications of UV de-
tector are research-missile warning systems, high tempera-
ture flame detection, air quality monitoring, gas sensing [4],
accurate measurement of radiation for the treatment of UV
irradiated skin. High excitonic binding energy (approx.
60 meV) of ZnO provides efficient excitonic emission at
room temperature used for laser with low thresholds even
at low temperature. In addition, the efficient blue-green
emission at room temperature, makes ZnO very important
for the fabrication of practical devices like light emission
diodes, laser diodes, and solar cells [5]. ZnO transparent
thin films transistors (TTFTs) are a very recent develop-
ment also [7].
In recent years, the silica-based nanocomposites are
considered the most challenging systems for the quantum
confinement of semiconductive nanocrystallites, for a bet-
ter control of shape, size and properties [5]. Preparation of
ZnO nanoparticles confined in silica matrix is investigated
for high UV luminescence at room temperature [8]. Zinc
oxide known as a luminescent material is today used for
various applications such as vaccum fluorescent displays
[8, 9]. Mesoporous silica-based nanocomposite films are
very important materials also for gas-sensing applications.
The well controlled mesoporosity of silica matrix provides
a path for the gas molecules to reach the functional oxide
nanoparticles and the films show reversible change in re-
sistance and/or optical transmittance in the VIS-NIR range
when exposed to gases. Various porous metal oxide-silica
nanocomposite films with enhanced gas-sensitive proper-
ties have been proposed for gas sensors [5, 6, 8, 9].
Different chemical methods are investigated for the
generation of different nanoparticles inside a SiO2 matrix,
which can act as a capping agent for the semiconductive
nanoparticles [10–14]. The sol–gel method is one of the

© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

 pss a
solidi
physica
status
2076 V. Musat et al.: ZnO/SiO2 nanocomposite thin films by sol – gel method
most efficient chemical method for the preparation of ox-
ide and silica based nanocomposite thin films and repre-
sents a simple and low-cost processing alternative to the
vacuum deposition techniques [1–3, 5–8]. This paper pre-
sents some results on ZnO/SiO2 nanocomposite thin films
obtained by sol–gel method using three different ways of
the sol preparation. The optical, electrical and UV/gas
sensing properties of ZnO/SiO2 nanocomposite films are
discussed in relation with their crystalline structure, mor-
phology and zinc-doping states.
2 Experimental
2.1 Films preparation The sols used for films depo-
sition were prepared starting from a pre-hydrolyzed tetra-
ethoxysilicate (TEOS) ethanolic solution, to which cetyl
trimethylammonium bromide (CTAB) as surfactant was
added. The molar ratio of TEOS:C2H5OH:H2O:HCl:
CTAB was 1:25:7:0.14:0.05. Starting from this solutions,
the following sols were prepared by:
– direct dissolution of zinc acetate solid salt (Sol I);
– adding of zinc acetate sol (preparation in [15] (Sol II);
– adding of ZnO nanoparticles prepared according to
– [16] (Sol III).
The thin films (two layers) were deposed on Corning
1737 glass substrate by dip-coating technique with differ-
ent withdrawal speed (Table 1).
After each layer deposition, the gel films were dried
for 30 min at 70 °C and then pre-heated in air for 10 min at
350 °C. The stabilized films were then post-heat treated
1 hour in air at 500 °C for crystallization. Zn concentration
on the surface of the films prepared using sol I, II and III
were measured by X-ray photoelectron spectroscopy
(XPS) technique (Table 1).
2.2 Films characterization The X-ray diffraction
(XRD) patterns of the samples were recorded at room tem-
perature using a Rigaku diffractometer (model RAD IIA),
with Cu Kα radiation. The morphology on the surface of
the films was analyzed using the atomic force microscopy
(AFM). Tapping mode AFM experiments were performed
in a Nanoscope IIIa Multimode AFM microscope (Digital
Instruments, Veeco). Commercial etched silicon tips with
typical resonance frequency of ca. 300 Hz (RTESP,
Veeco) have been used as AFM probes. The optical trans-
mittance was measured using a UV–VIS-NIR double beam
Table 1 Experimental conditions and zinc concentration.
sample sol number film deposition chemical composition*
cod (cm/min)
S1 I 20 16.4
S2 I 30
S3 II 20 31.0
S4 II 30
S5 III 20 4.6
S6 III 30
*
Zn concentration (wt%) calculated from XPS data.
© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
spectrophotometer (Shimadzu, UV-3100 PC) in the wave-
length range 200 nm to 2500 nm. From the optical meas-
urements, the optical energy gap, Eg, was calculated as-
suming a direct transition between the edges of the
valence and the conduction bands, for which the varia-
tion in the absorption coefficient, α, with photon energy,
hν, is given by the equation:
(αhν)2 = B(hν – Eg) . (1)
2
By plotting (αhν) versus hν, and extrapolating the linear
region of the resulting curves, Eg was obtained.
2.3 Sensing test The photoreduction – gas (ozone)
reoxidation sensing behavior of thin films was investigated
by electrical measurements at room temperature, which were
carried out in a special chamber presented elsewhere [17].
The conductivity of the films increases by photoreduction,
when the films were directly irradiated in vacuum by the UV
light of a mercury pencil lamp with an average intensity of
4 mW/cm2 at 254 nm for 20 min and then decreases when
the films are exposed in an ozone atmosphere. First the film
was exposed at UV light irradiation and after that, the cham-
ber was backfilled with oxygen at a pressure of 600 Torr and
an UV lamp was used to produce ozone (the films are
shielded from the lamp). The films were maintained for
40 min in this ozone atmosphere, when the re-oxidation of
zinc oxide crystallites results in a decrease of the film con-
ductivity. An electric field of 50 V/cm was applied to the
film sample and the electrical current was measured.
3 Results and discussion Thicknesses of about 190
and 290 nm have been measured from the ellipsometric
data (not presented in this paper) for the crystallized films
(two layers) deposed with withdrawal speed of 20 cm/min
and 30 cm/min, respectively. The XPS photoelectron spec-
tra of Zn 2p3/2 region (Fig. 1) show Zn–O–Si
S1
10000 S2
S3
S4
8000 S5
Intensity, a.u.
S6
6000
4000
2000
1032.0 1027.0 1022.0 1017.0 1012.0
Binding Energy, eV
Figure 1 (online colour at: www.pss-a.com) Zn 2p XPS photo-
electron peaks.
www.pss-a.com
 Original
Paper

phys. stat. sol. (a) 205, No. 8 (2008) 2077

8 00 0

S1
S2
6 00 0 S3
S4
Intensity, a.u.

S5
S6
4 00 0

2 00 0

0
9 75 98 0 98 5 99 0 9 95 1 00 0
B in d in g En ergy , e V

Figure 2 (online colour at: www.pss-a.com) Zn LMM Auger

peaks for samples S1 – S6.

bonds in all the films. Zn peaks are narrower for the sam-
ples S3 and S4 than for the other samples, indicating a
smaller dispersion of neighborhoods for ZnO. The most
charged samples S1–S2 and S5–S6 are more insulating
and the ones where the hole trapping is more efficient.
The XPS quantitative results show that samples S3 and
S4 are also the samples where the zinc oxide embedding
was more efficient. The Zn LMM Auger peaks (Fig. 2)
confirm and highlight the difference in binding energy
value for zinc atoms in the samples S3–S4 comparative
with the others films.
The XRD patterns of the thin films deposed using the
sol II at 20 cm/min and 30 cm/min (samples S3 and S4) are
presented in Fig. 3 and show wurtzite tipe ZnO nanocrys-
talline phase with preferential (100) crystalline orientation,
embedded in an amorphous silica matrix. The thin films Figure 4 (online colour at: www.pss-a.com) AFM images on the
deposed at lower withdrawal speed or using the sols I and films surface of samples S3 and S4 (scale in nm).
III didn’t show well defined peaks of ZnO phase.
The AFM images (Fig. 4) show cracks free mesopor- The optical transmittance spectra (Fig. 5) show a good
ous films morphology with quasi-regularly shaped grains transmittance (between 85% and 95%) in the UV–VIS and
and regular roundish pores size; higher withdrawal speed a NIR regions and a significant blue shift of the absorption
little higher grain size and pore size. bands.
Figure 6 presents the optical energy gap (Egap) calcula-
700 tions for the thin films prepared using the three mentioned
ZnO (100)

600 S2
100
500
200
S4 S6
80 S4
Transmitance(%)
Intensity (a.u.)

Intensity (a.u.)

400 20
S3 60
Transmitance (%)

15
300
ZnO (200)

100
40 10

200 S6
5 S2
20 S4
30 31 32 33 34 35
0
100 2Theta (deg.)
2 00 22 0 24 0 26 0 28 0 3 00 3 20
0 W a ve le ng th (n m )
0
20 25 30 35 40 45 50 55 60 65 70 0 500 1000 1500 2000 2500
2Theta (deg.) Wavelenght(A)

Figure 3 XRD patterns of the samples S3 and S4. Figure 5 Optical transmittance spectra of samples S2, S4, S6.

www.pss-a.com © 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

 pss a

solidi
physica

status
2078 V. Musat et al.: ZnO/SiO2 nanocomposite thin films by sol – gel method

1.00¥1016 1 0 −5

S2 S2
8.00¥1015

Conductivity (Ω cm )
-1
1 0 −6
2

6.00¥1015
(ahn) (eV/cm)

-1
1 0 −7
4.00¥1015
2

2.00¥1015 1 0 −8

0
Eg=4.859 eV 1 0 −9
-2.00¥1015 2 00 0 4 0 00 6 0 00

0 1 2 3 4 5 6 T im e (min )
hn (eV)
8.00¥1015 1 0 −5
2
(ahn) (eV/cm)

S4 S4

Conductivity (Ω cm )
-1
6.00¥10 15
1 0 −6
2

-1
4.00¥1015
1 0 −7

2.00¥1015
1 0 −8
0
Eg=4.264 eV
1 0 −9
-2.00¥1015 0 20 00 4 0 00 6 00 0 8 00 0
0 1 2 3 4 5
T im e (min )
hn (eV)
8.00 ¥1015 1 0 −5
2
(eV/cm)

S6 S6
Conductivity (Ω cm )

6.00 ¥1015
-1

−6
10
-1
2
( ah n )

4.00 ¥1015
1 0 −7

2.00 ¥10 15

0
-2.00 ¥1015
0 1
Figure 6 Band gap calculation for samples S2, S4 and S6.
sols and deposed at 30 cm/min. Very small differences (de-
crease) have been observed for the Eg values of the films de-
posed at 20 cm/min. The band gap values are very high at
room temperature. The blue shift of the band gap from that of
bulk ZnO (3.3 eV) towards 4.26 eV for sample S4, confirm
that an effective capping of the ZnO nanoparticles in the
SiO2 matrix of the nanocomposite films have been obtained.
This significant blue shift of the ZnO band gap energy
demonstrating the quantum confinement effect of ZnO
nanoparticles in the silica or alumina host matrix has been
also reported by other authors [8]. Much smaller zinc oxide
nanocrystallites (clusters) and/or higher cross links Zn–
O–Si binder energy [10, 12] for the samples S1–S2 and
S5–S6 (Fig. 1) led to a higher carrier confinement for
these silica-based nanocomposite films and a much higher
blue shift of the band gap energy (S2 and S6 in Fig. 6).
Figure 7 presents the UV photoreduction-ozone reoxi-
dation cycles of the film prepared using the sols I, II and III,
1 0 −8
E g=5.032 eV
1 0 −9
20 00 400 0 60 00
2 3 4 5 6 7
h n (eV) Tim e (min)
Figure 7 Photoreduction – ozone reoxidation cycles for samples
S2, S4 and S6.
deposed at withdrawal values of 30 cm/min. All the films
have a reversible room temperature behavior during UV ir-
radiation followed by re-oxidation in ozone atmosphere.
From Fig. 7, it can be noticed a slight increase in
changing rezistivity response to over three orders of mag-
nitude, once with the growth of zinc oxide concentration
(sample S4).
4 Conclusion Zinc oxide silica-based nanocomposite
mesoporous thin films were prepared by sol–gel method
using three different ways of sol preparation: direct disso-
lution of zinc acetate salt (I), adding of zinc acetate sol (II)
and adding of ZnO nanoparticles (III) to a pre-hydrolyzed
tetraethoxysilane (TEOS) ethanolic solution, to which cetyl
trimethylammonium bromide (CTAB) as surfactant was
added. For each method, the stability of the resulted sol led
to different zinc concentration. Values of 16.4, 31.1 and

© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.pss-a.com


phys. stat. sol. (a) 205, No. 8 (2008)
4.6 wt% were obtained for the thin films prepared by meth-
ods I, II and III, respectively. In the films with larger con-
centration of zinc (31.1 wt%), (100) oriented wurtzite type
ZnO nanocrystallites were detected. These are the samples
where the zinc oxide embedding was more efficient.
All the films are high transparent. Optical transmit-
tance between 85% and 95% within the visible and near IR
wavelength region has been obtained. The blue shift of the
band gap towards 4.2 eV for samples prepared using sol II,
confirm that an effective capping of the ZnO nanoparticles
in the SiO2 matrix of the nanocomposite films have been
obtained. Much smaller zinc oxide nanocrystallites (clus-
ters) and/or higher cross links Zn–O–Si binder energy for
the samples prepared using sols I and III led to a higher
carrier confinement for these nanocomposite films and a
much higher blue shift of the band gap (4.85 and 5.03 eV,
respectively).
All the investigated films revealed a consistent reversi-
ble room temperature change in electrical resistance during
UV irradiation followed by re-oxidation in ozone atmos-
phere. The magnitude of the sensing response (resistivity
variation during the photoreduction – reoxidation cycles) is
slightly depending on zinc concentration. Higher zinc con-
centration higher photoreduction – oxidation response
(three order of magnitude in resistivity variation).
Acknowledgement This work was supported by NATO
Expert Visit Grant PDD(CP)-CBP.EAP.CLG 982806 /2007.
References
[1] S. Bandyopadhyay, G. K. Paul, and S. K. Sen, Sol. Energy
Mater. Sol. Cells 71, 103 (2002).
www.pss-a.com
Original
Paper
2079
[2] S. Fujihara, C. Sasaki, and T. Kimura, Appl. Surf. Sci. 180,
341 (2001).
[3] C. R. Ding, S. W. Li, and H. Z. Wang, Appl. Phys. Lett. 90,
241918 (2007).
[4] W. Shen, Y. Zhao, and C. Zhang, Thin Solid Films 483, 382
(2005).
[5] S. Chakrabarti, D. Das, D. Ganduli, and S. Chaudhuri, Thin
Solid Films 441, 228 (2003).
[6] L. Armelao, D. Barreca, S. Gross, A. Martucci, M. Tieto,
and E. Tondello, J. Non-Cryst. Solids 293 – 295, 477 (2001).
[7] B. J. So, Noris, J. Anderson, J. F. Wager, and D. A. Keszler,
J. Phys. D, Appl. Phys. 36, L105 (2003).
[8] S. Chakrabarti, D. Ganguli, and S. Chaudhuri, phys. stat. sol.
(a) 201, 2134 (2004).
[9] W. C. Chen, Mater. Lett. 59, 1239 (2005).
[10] L. I. Burova, D. I. Petukhov, A. A. Eliseev, A. V. Lukashin,
and Yu. D. Tretyakov, Superlattices Microstruct. 39, I257
(2006).
[11] D. Haranath, N. Bhalla, H. Chander, R. Meenakshi Kar, and
R. Kishore, J. Appl. Phys. 96, 6700 (2004).
[12] D. M. Abdullah, S. Shibamoto, and K. Okuyama, Opt. Ma-
ter. 26, 95 (2004).
[13] B. Yao, H. Shi, H. Bi, and L. Zang, J. Phys.: Condens. Mat-
ter 12, 6265 (2000).
[14] Y. Y. Peng, T. E. Hsieh, and C. H. Hsu, Nanotechnology
17, 174 (2006).
[15] V. Musat, B. Teixeira, E. Fortunato, R. C. C. Monteiro, and
P. Vilarinho, Surf. Coat. Technol. 180/181, 659 (2004).
[16] P. Elumalai, H. N. Vasan, and N. Munichandraiah, Mater.
Res. Bull. 39, 1895 (2004).
[17] A. Pimentel, E. Fortunato, A. Gonçalves, A. Marques,
H. Águas, L. Pereira, I. Ferreira, and R. Martins, Thin Solid
Films 487, 212 (2005).
© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim