Thin Solid Films 518 (2010) 1774–1778

Contents lists available at ScienceDirect

Thin Solid Films

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / t s f

Optical properties of CdSe and CdO thin films electrochemically prepared

R. Henríquez a, P. Grez a, E. Muñoz a, H. Gómez a, J.A. Badán b, R.E. Marotti b,⁎, E.A. Dalchiele b
a
Instituto de Química, Facultad de Ciencias, Pontificia Universidad Católica de Valparaíso, Casilla 4059, Valparaíso, Chile
b
Instituto de Física, Facultad de Ingeniería, Universidad de la República, CC30, Montevideo, Uruguay

a r t i c l e i n f o a b s t r a c t

Available online 19 September 2009 The present work comparatively studies the optical properties of CdO against CdSe samples potentiosta-
tically electrodeposited onto fluorine-doped tin oxide/glasses (FTO). The films were prepared by
Keywords: electrochemical reduction processes in dimethyl sulfoxide (DMSO) solution. The optical properties are
Optical transmittance studied by transmittance measurements and diffuse reflectance spectra (DRS). Structural properties of the
Electrodeposition films were also studied by X-ray diffraction spectroscopy (XRD). For CdSe the direct bandgap energy
Absorption coefficient
obtained from both transmittance measurements and first derivative peak position of the DRS is 1.69 ±
Bandgap
0.01 eV. In the case of CdO both direct and indirect gaps were found. Careful deconvolution of the spectra
allows assigning the direct bandgap energy at 2.52 eV. A first indirect bandgap appears at 2.03 eV, while it is
possible to have a second one at 0.46 eV.
© 2009 Elsevier B.V. All rights reserved.

1. Introduction effect, to have low resistivity value, as required for transparent con-
ductive oxides.
During the past years there has been an increasing interest in Still, the optical properties of CdO are complicated because there
semiconductor oxides and II–VI binary semiconducting alloys for thin have been reports of different indirect gaps with energy values at
film solar cells. The former ones, semiconductor oxides like ZnO, are 0.55 eV [17,18], ~0.8 eV [19,20], 1.1 eV [19], 1.36 eV [21], 1.47 eV [19]
used as transparent conductive windows [1] or as buffer layers [2]. On and from 2.1 eV [8] to 2.3 eV [11]. The correct understanding of CdO
the other hand, Cd based II–VI binary alloy semiconductors are very optical absorption edge is of major importance for any optoelectronic
interesting for their application in efficient multilayer thin film solar application of this material. In the present work, the optical properties of
cells [3]. CdTe, CdSe and CdS are all direct bandgap semiconductors CdO thin films are comparatively studied against CdSe thin films
with reported room temperature bandgap energies at 1.49 eV, prepared by electrochemical deposition. This method is chosen because
1.74 eV, and 2.42 eV, respectively [4,5]. These values depend slightly it is a relative low cost technique capable of large area depositions.
on crystal structure, being 1.49 eV for CdTe in cubic (zincblende),
while the later values corresponding to hexagonal (wurtzite) CdS and 2. Experimental
CdSe. With these bandgap energies it is possible to efficiently absorb
light from the IR to the visible spectral region. The semiconductor CdO and CdSe thin films were electrochemically (potentiostatically)
oxide corresponding to this family is CdO. CdO was not so well studied prepared by employing 5 mM CdCl2 and 0.1 M LiCl in dimethyl sulfoxide
as other semiconductor oxides (e. g. ZnO which has a bandgap energy (DMSO) solution by electrochemical reduction process consisting of
at the UV: 3.3 eV [6]), mainly because its direct bandgap energy is not saturated molecular oxygen (O2), for CdO, and 1.2 mM of selenourea
so high. The direct band gap energy of CdO is usually reported to be (Se = C(NH2)2), for CdSe. The electrodeposition of films was carried out
between 2.25 to 2.9 eV [7–13]. The blue shift of this value may be at 120 °C in a conventional three electrode cell. Pt wire was used as
attributed to band filling effect generated by excess free carriers counter electrode; saturated Ag/AgCl was used as reference and
through Moss–Burstein effect [9,10,13–15]. However there has been fluorine-doped tin oxide/glass (FTO) as working electrode at constant
recently an increased interest in this material because it has high potentials (Ed) of −0.80 V and −1.20 V, respectively. Only the CdO films
mobility values [9,11,16]. This high mobility is needed, together with were thermally treated at 350 °C for 2 h under oxygen in a Lindberg/
the high free carrier concentration that generates Moss–Burstein Blue M Tube Furnace.
Structural properties of the films were studied by powder X-ray
diffraction (XRD) spectroscopy with a Philips PW3710 diffractometer
using CuKα radiation. The optical properties of the samples were
⁎ Corresponding author. Instituto de Física, Facultad de Ingeniería, Universidad de la
República, Julio Herrera y Reissig 565, CC30, CP 11000, Montevideo, Uruguay. Tel.: +598 2
studied by optical transmittance spectra measurements and diffuse
711 09 05; fax: +598 2 711 16 30. reflectance spectroscopy at the visible and near infrared region. A
E-mail address: khamul@fing.edu.uy (R.E. Marotti). tungsten incandescent filament lamp was used as light source and

0040-6090/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.tsf.2009.09.030
 R. Henríquez et al. / Thin Solid Films 518 (2010) 1774–1778 1775

an Ocean Optics S2000 fiber optic spectrometer (with a Sony ILX511

Linear CCD Array and a 100 µm core diameter fiber acting as input
slit) recorded the optical spectra between 350 and 1000 nm. An FTO/
glass substrate was used as a reference. The diffuse reflectance was
detected normal to the surface with light impinging the sample at
quasi-normal incidence [22]. All measurements were done at room
temperature.

3. Results and discussion

In order to study the structural properties of the CdSe and CdO

films, i.e.: to investigate the crystallographic phase, the overall
crystalline quality, and the possible texture of those electrochemically Fig. 2. Transmittance spectra of a CdO sample (short dash), CdSe sample (dashed), and
bare FTO/glass substrate (full lime).
grown thin films, X-ray diffraction experiments have been carried
out. Fig. 1 shows typical XRD patterns for the CdSe and CdO electro-
deposited thin films, and for comparison their respective JCPDS pat-
terns are also included [23,24]. The diffractograms revealed a number the visible region although it has a transmittance of the order of 20%
of peaks corresponding to CdSe, CdO and FTO substrate layer. It is close to 600 nm, corresponding to its color. The CdSe film transmit-
known that the bulk CdSe has a highly stable hexagonal (wurtzite) tance is never higher than 80% even at the IR. The CdO film has a
phase from room temperature to melting point, while cubic CdSe transmittance higher than 90% in the IR although it absorbs light in
exists at low temperature and it is the metastable phase [25]. In the blue–UV region, corresponding to its yellow color. The thickness
our case, the diffractogram of the CdSe film (Fig. 1a) shows a charac- of the FTO film was on the order of 1 µm as specified by the provider;
teristic polycrystalline hexagonal wurtzite structure with a slight meanwhile the CdSe and CdO films were estimated from AFM to be
(002) crystallographic preferred orientation. On the other hand Fig. 1b 0.26 and 0.55 µm, respectively. Therefore all spectra have interference
shows the XRD pattern of the electrodeposited CdO films exhibiting maxima and minima, i.e. the interference patterns were already
a polycrystalline cubic rock salt structure, with a very high (200) present at the FTO film of the substrate. After growing the films these
preferred orientation. maxima and minima change position because of differences in the
The resulting samples are homogeneous to the naked eye. The effective optical path. In spite of that it can be seen that the CdSe and
CdSe samples have a deep reddish color while the CdO ones have a CdO film spectra have maxima close to FTO/glass minima, at least in
light yellow color [9]. Their comparative optical transmittance spectra their transparency region, so the interference effects are enhanced
T are shown in Fig. 2. In this figure the transmittance of the FTO/glass because of calculating transmittance relative to FTO/glass bare
substrate is an absolute measurement (light source spectrum used as substrate. This complicates the study of the optical properties of the
a reference) while for both CdO and CdSe film, the FTO/glass substrate films, especially in the case of CdO film, which has already a very reach
was used as reference. The substrate has a transmittance less than 80% optical structure. For that reason the optical properties of CdSe are
because it has a white (“milky”) appearance to the naked eye (light studied first for comparison.
dispersion of the bare substrate) [26]. The CdSe film absorbs light in
3.1. Optical properties of CdSe

The optical properties of CdSe samples are studied in Fig. 3. Fig. 3a

shows typical (α × hν)2 plots for direct energy gap determination
[4,27,28], where α ≈ − lnT is the absorption coefficient and hν is the
photon energy. The bandgap energy is obtained by extrapolating a
linear fitting to the zero absorption line. In Fig. 3a, the dashed line
corresponds to the CdSe spectra of Fig. 2 while the dashed dotted
line corresponds to the transmittance of the whole CdSe/FTO/glass
samples (i.e. without taking a bare substrate as reference). The
bandgap energy values obtained from these cases are 1.68 eV and
1.64 eV, respectively. These values are smaller than the accepted value
for CdSe with wurtzite structure (in less than 0.1 eV), which is 1.74 eV
[4,5,28]. From Fig. 3a, the plot corresponding to the whole CdSe/FTO/
glass substrate is more linear than the one corresponding just to the
CdSe film (due to the interference effects already present in the sub-
strate). In spite of that the corresponding value may have a larger
uncertainty because of the IR transmittance of the whole sample,
which results in uncertainty of the zero line for extrapolating the
linear fitting. For that reason the following procedure was developed
for improving the bandgap energy determination.
Close to the transparency region the inverse transmittance 1/T vs.
wavenumber 1/λ is expected to have an oscillatory behavior [27]. The
onset of absorption complicates this oscillatory behavior. Subtracting,
from the 1/T vs. 1/λ measured curve, a smoothed curve of the same
spectra, approximately eliminated the effect of absorption. Another
equivalent approach is to subtract a low order polynomial fitting. The
Fig. 1. X-ray diffraction patterns of (a) CdSe and (b) CdO thin films electrodeposited
onto SnO2:F substrates. Hexagonal wurtzite CdSe and cubic CdO JCPDS patterns are also
result is an oscillatory function that is not a harmonic function due to
shown for comparison (black lines). (⁎, indicates the peaks originated from the SnO2:F the effect of the interference at the substrate. In spite of that, a cosine
substrate). function is adjusted to this oscillatory curve close to the absorption
1776 R. Henríquez et al. / Thin Solid Films 518 (2010) 1774–1778

relative height and width of this peak, and the step like behavior of the
corresponding feature in the reflectance spectrum, allows assigning it to
the desired absorption edge. Therefore the best value obtained for CdSe
film bandgap energy is 1.69 eV, i.e. the one from either methods of
Fig. 3b and c, which is close to the expected value.

3.2. Optical properties of CdO

The optical properties of CdO samples are studied in Fig. 4. Fig. 4a

shows typical (α × hν)2 plot for direct energy gap as in Fig. 3a. The
absorption coefficient used in Fig. 4a is obtained from the short dash
of Fig. 2. The bandgap energy obtained in this case was 2.49 eV, which
is in the region of direct bandgap energy usually reported for CdO,
between 2.25 to 2.9 eV [7–13]. The energy span of this value is usually
assigned to small departures from stoichiometry [10], which
generates defect states and excess carriers. The excess carriers have
two different and opposing effects: first, the shrinkage of the true
bandgap due to many body effects [15]; and second, a band filling of
the low energy states at the conduction band [10] by excess free
carriers due to Pauli exclusion principle. Both effects taken into
account generate a net blue shift of the absorption edge known as
Moss–Burstein effect [9,10,14,32]. Therefore the value of 2.49 eV
obtained in Fig. 4a is in the expected region, close to most reported
values [8,10], corresponding to a small departure from stoichiometry.
Still, it can be seen in Fig. 4a (and Fig. 2), and will be studied in detail
in what follows, that in the 2.2–2.5 eV region (i.e. ~ 565–495 nm),
there is a sub-bandgap absorption. No interference effects were seen
in the bandgap energy determination of Fig. 4a. This is due to the fact
that the transmittance of CdO film in the transparency region (>90%)
is higher than the one of CdSe film (<80%).
Diffuse reflectance measurements of the whole CdO/FTO/glass
sample were also carried out and the results are shown in Fig. 4b. It
can be seen again that there are strong oscillations due to interference
and a shoulder smoother than the one observed in the case of CdSe
(see Fig. 3c). The first derivative is also shown in Fig. 4b, whose
strongest peak is at 508 nm which corresponds to a bandgap energy of
2.44 eV, i.e. 0.05 eV smaller than the one obtained at Fig. 4a. This small
discrepancy is expected in bandgap energy determination from
reflectance measurements [29]. However, comparing Fig. 4b with
Fig. 3c, it is seen that in the case of CdO sample the peak is not so
Fig. 3. Direct bandgap energy determination for CdSe samples: a) (α × hν)2 plots and sharp. A close peak at 540 nm indicates that interference may be
linear fittings for CdSe film (dashed line) and whole CdSe/FTO/glass (dot–dashed line). affecting the measurement from this method in the present case.
b) (α × hν)2 plots and linear fitting for CdSe film after analytical elimination of
interference pattern in transmittance spectra (see inset), c) Diffuse reflectance spectra
Procedures to analytically eliminate the interference in the region
and its first derivative. close to the absorption edge, as was done in the inset of Fig. 3b, were
tried without success. Therefore another structure is present in the
absorption edge, apart from direct bandgap absorption and interfer-
edge (i.e. ~ 750 nm, which is near the previously obtained bandgap ence effects.
energy). The adjusted cosine can then be used to eliminate the oscil- Fig. 4c shows typical Tauc (α × hν)1/2 vs. hν plot for indirect
latory part of the 1/T vs. 1/λ measured curve close to the absorption bandgap energy determination [4,27,33]. Once again the bandgap
edge, therefore one can recalculate the transmittance spectra without energy corresponding to an indirect transition is obtained by extra-
interference pattern in the desired region. From that transmittance polating a linear fitting to the zero absorption line. In Fig. 4c the short
spectrum without oscillations a better bandgap energy value can be dashed line was calculated directly from α ≈ − lnT with T the
estimated. The result of this process is shown in Fig. 3b. The inset transmittance spectra for CdO sample in Fig. 2. The influence of the
shows the transmittance without interference close to 750 nm. The interference patterns is again evident in this figure, as many
bandgap energy value obtained from this procedure is 1.69 eV. oscillations appear in the low energy region of this curve — which
To check the previous procedure, and bandgap energy determina- corresponds to the transparency region of transmittance. If just one
tion, diffuse reflectance measurements of the CdSe/FTO/glass sam- linear fitting is tried in the region between 1.2 and 2.4 eV (i.e. below
ple were carried out using a bare FTO/glass substrate as a reference previously obtained value of direct bandgap energy) indirect bandgap
[22,29,30]. Fig. 3c shows the result of this measurement and its first energy of 0.9 eV is obtained (see inset of Fig. 4c). This value is close to
derivative. The reflectance spectrum also has an oscillatory behavior, previous reports of indirect bands in CdO with bandgap energy at
although a step like shoulder appears close to 750 nm. The peak of the 0.8 and 1.1 eV [19,20]. Therefore it can be concluded that the below
first derivative corresponding to this shoulder is another measure- direct bandgap absorption is due to these close indirect band transi-
ment of the direct bandgap energy [29,31]. The value obtained by this tion. Following this line of argumentation the departure of the curve
method is again 1.69 eV. Although in the case of strong interference the of the inset at Fig. 4c from the linear fitting can be assigned to the
peak derivative may be shifted by the peaks arising from the inter- interference. However, the previous studies done on CdSe for ana-
ference oscillatory behavior, in the present case it does not happen. The lytically eliminating the interference maxima and minima, which
 R. Henríquez et al. / Thin Solid Films 518 (2010) 1774–1778 1777

Fig. 4. Bandgap energy determination for CdO samples: a) Direct gap determination as in Fig. 3a with α = −lnT from short dots of Fig. 2. b) Diffuse reflectance spectra and its first
derivative. c) Indirect gap determination from (α × hν)1/2 plots: Short dash is as before and dotted dash is after background subtraction. Inset shows a linear fitting in an extended
region. d) Measured absorption coefficient and superposition of results from previous determinations.

could not be successfully applied to CdO sample, prevent us from Having assigned the below direct bandgap absorption to two
extracting this conclusion. Careful analysis of the short dashed curve different indirect-like absorptions, with bandgap energies at 2.03 eV
in Fig. 4c (and the corresponding spectrum in Fig. 2) allows separating and 0.46 eV, the direct bandgap energy can be corrected. If the
the below direct bandgap energy region in two different ones: first indirect absorption is now considered as a background (new zero line
there is a linear region between 2.1 and 2.4 eV (i.e. ~590–520 nm), absorption for the direct band absorption) corrected direct bandgap
and a second oscillating one below 2.1 eV (i.e. the transparency energy can be obtained. This is done in the inset of Fig. 4d and a
region: high transmittance in Fig. 2). A linear fitting for the former one bandgap energy of 2.52 eV was obtained. Therefore it can be
is possible (not shown) giving indirect bandgap energy at 1.85 eV. concluded that CdO transmittance of Fig. 2 show a direct absorption
This is again close to other reported values of indirect bandgap edge with bandgap energy at 2.52 eV and an indirect absorption edge
energies in CdO close to 2 eV [8]. However, in the present case, the with energies at 2.03 eV. The transparency region has a low
value of 1.85 eV has a similar uncertainty than the one for the direct absorption that can correspond to another indirect and very weak
bandgap energy of CdSe discussed in relation to Fig. 3a, originated in absorption edge at 0.46 eV.
the position of the zero absorption line. So, in order to eliminate this In Fig. 4d, the measured absorption coefficient is shown together
uncertainty, the following procedure was applied; which was already with the contribution of each of the previous: the short dashed line is
successfully used for nanometric sized ZnO films [34]. α ≈ − lnT, the measured absorption; the thick full line is the
In the short dashed curve of Tauc plot of Fig. 4c, a linear fitting is absorption of indirect band nature with bandgap energy at 0.46 eV;
done for the below both (indirect and direct) gaps previously the thin full line considers both contributions of the previous one plus
obtained. The absorption coefficient obtained from this fitting is indirect band absorption with bandgap energy at 2.03 eV — it tends
considered as background absorption (i.e. corresponding to the zero asymptotically to the previous one; and the dashed line is direct
line absorption) and subsequently subtracted from the measured absorption which superimposed to the previous one gives the dotted
absorption coefficient. A new analysis can be done in this corrected line, which considered all three contributions. The adjusted curve
absorption. The linear fitting for this background absorption is shown follows very close the measured one. The major discrepancies are
in Fig. 4c and corresponds to indirect bandgap energy of 0.46 eV. Care present at the low energy region due to the interference at the
must be taken in really assigning a gap to this value for two reasons: transparency region, and in the onset of direct absorption. This
first, it was observed previously that these very low absorption values discrepancy at this point can be originated both in the band filling
can be due to disorder present in the sample [22,33,34] and not a true effect due to Moss–Burstein effect [32] or band tailing (Urbach tail)
interband transition; and second, the extrapolated value is in this case usually assigned to local mechanical stress produced by impurities
in the IR region away from the measured spectra region which just and structural defects [4,35].
reaches 1000 nm (1.24 eV). In spite of that there have already been
reports of indirect bands at this IR region for CdO [17,18]. The dash 4. Conclusions
dotted line in Fig. 4c is the Tauc plot for the corrected absorption
coefficient (subtraction of the background absorption from previous CdSe and CdO films were prepared by electrochemical reduction
linear fitting). Next, corrected indirect bandgap energy can be processes in dimethyl sulfoxide solution. X-ray diffraction measurements
obtained from the linear fitting between 2.1 to 2.4 eV. The resulting show that the crystalline phases present in the samples are hexag-
value obtained in Fig. 4c for this indirect bandgap energy is 2.03 eV. onal wurtzite structure for CdSe and cubic rock salt for CdO. The
1778 R. Henríquez et al. / Thin Solid Films 518 (2010) 1774–1778

optical properties of CdSe films are those of a typical direct gap [9] X. Li, D.L. Young, H. Moutinho, Y. Yan, C. Narayanswamy, T.A. Gessert, T.J. Coutts,
Electrochem. Solid-State Lett. 4 (2001) C43.
semiconductor in spite of that they show strong interference fringes. [10] R. Jayakrishnan, G. Hodes, Thin Solid Films 440 (2003) 19.
These were already present at the FTO substrate and can be decon- [11] J. Santos-Cruz, G. Torres-Delgado, R. Castanedo-Pérez, S. Jiménez-Sandoval, O.
voluted from the measurement. The direct bandgap energy obtained Jiménez-Sandoval, C.I. Zúñiga-Romero, J. Márquez-Marín, O. Zelaya-Angel, Thin
Solid Films 493 (2005) 83.
for CdSe samples from both transmittance measurements and the [12] A. Seshadri, N.R. Tacconi, C.R. Chenthamarakshan, K. Rajeshwar, Electrochem.
position of the peak of first derivative of DRS is 1.69 ±0.01 eV, in Solid-State Lett. 9 (2006) C1.
good agreement with that of bulk CdSe. In the case of CdO films the [13] L. Wang, Y. Yang, S. Jin, T.J. Marks, Appl. Phys. Lett. 88 (2006) 162115.
[14] B.E. Sernelius, K.F. Berggren, Z.C. Jin, I. Hamberg, C.G. Granqvist, Phys. Rev., B 37
deconvolution process is not possible showing that CdO has a more (1988) 10244.
complicated band edge. After a careful analysis of this absorption [15] Y. Dou, T. Fisklock, R.G. Egdell, D.S.L. Law, G. Beamson, Phys. Rev., B 55 (1997)
edge both direct and indirect gaps were found. The direct bandgap R13381.
[16] X. Li, Y. Yan, A. Mason, T.A. Gessert, T.J. Coutts, Electrochem. Solid-State Lett. 4
for CdO is at 2.52 eV, while the first indirect bandgap is at 2.03 eV.
(2001) C66.
Another indirect bandgap close to 0.46 eV could also be assigned. [17] X. Liu, C. Li, S. Han, J. Han, C. Zhou, Appl. Phys. Lett. 82 (2003) 1950.
[18] H. Köhler, Solid State Commun. 11 (1972) 1687.
Acknowledgments [19] F.P. Koffyberg, Phys. Rev. 13 (1976) 4470.
[20] A.B.M.A. Ashrafi, H. Kumano, I. Suemune, Y.-W. Ok, T.-Y. Seong, Appl. Phys. Lett. 79
(2001) 470.
The authors acknowledge FONDECYT-CHILE support (N° 1060442) [21] X. Wu, R. Wang, B. Zou, P. Wu, L. Wang, J. Xu, W. Huang, Appl. Phys. Lett. 71 (1997)
and the project Incentive to the International Cooperation FONDECYT 2097.
[22] S.R. Johnson, T. Tiedje, J. Appl. Phys. 78 (1995) 5609.
(N° 7070037). The authors acknowledge especially the Bicentennial [23] Powder Diffraction File, Joint Committee for Powder Diffraction Studies (JCPDS)
Project of Science and Technology project of the Chile Government File No. 08-0459 (hexagonal wurtzite CdSe), 1999.
(N° PSD 82) and the Electrochemical Group Project (037.108.2008). [24] JCPDS File No. 05-0640 (cubic CdO), 1999.
[25] O. Portillo-Moreno, R. Lozada-Morales, M. Rubín-Falfán, J.A. Pérez-Alvarez, O.
E. A. D., J. A: B and R. E. M. acknowledge the support received from Zelaya-Angel, L. Baños-López, O. Portillo-Moreno, R. Lozada-Morales, M. Rubín-
PEDECIBA-Física, and the CSIC (Comisión Sectorial de Investigación Falfán, J.A. Pérez-Alvarez, O. Zelaya-Angel, L. Baños-López, J. Phys. Chem. Solids 61
Científica) of the Universidad de la República, in Montevideo, Uruguay. (2000) 1751.
[26] D. Ramírez, D. Silva, H. Gómez, G. Riveros, R.E. Marotti, E.A. Dalchiele, Sol. Energy
J. A: B. also acknowledges ANII. Mater. Solar Cells 91 (2007) 1458.
[27] O. Stenzel, The Physics of Thin Film Optical Spectra: An Introduction, Springer-
References Verlag, Berlin, 2005.
[28] G. Riveros, J. Vásquez, H. Gómez, T. Makarova, D. Silva, R.E. Marotti, E.A. Dalchiele,
[1] T. Repmann, B. Sehrbrock, C. Zahren, H. Siekmann, B. Rech, Sol. Energy Mater. Sol. Appl. Phys., A 90 (2008) 423.
Cells 90 (2006) 3047; [29] R.E. Marotti, D.N. Guerra, C. Bello, G. Machado, E.A. Dalchiele, Sol. Energy Mater.
J. Hüpkes, B. Rech, O. Kluth, T. Repmann, B. Zwaygardt, J. Müller, R. Drese And, M. Solar Cells 82 (2004) 85.
Wuttig, Sol. Energy Mater. Sol. Cells 90 (2006) 3054. [30] R.E. Marotti, J.A. Badán, E. Quagliata, E.A. Dalchiele, Physica B 398 (2007) 337.
[2] M. Igalson, C. Platzer-Björkman, Sol. Energy Mater. Sol. Cells 84 (2004) 93. [31] G. Riveros, H. Gómez, R. Henríquez, R. Schrebler, R.E. Marotti, E.A. Dalchiele, Bol.
[3] J.D. Beach, B.E. McCandless, MRS Bull. 32 (2007) 225. Soc. Chil. Quím. 47 (2002) 411.
[4] J.L. Pankove, Optical Processes in Semiconductors, Dover Publications, New York, [32] I. Hamberg, C.G. Granqvist, K.F. Berggren, B.E. Sernelius, L. Engström, Phys. Rev., B
1971. 30 (1984) 3240.
[5] R.N. Bhargava (Ed.), Properties of Wide Bandgap II–VI Semiconductors, EMIS [33] J. Tauc, Mater. Res. Bull. 5 (1970) 721.
Datareviews Series No 17, INSPEC, London, UK, 1997. [34] C.D. Bojorge, H.R. Cánepa, U.E. Gilabert, D. Silva, E.A. Dalchiele, R.E. Marotti,
[6] Ü. Özgür, Y.I. Alivov, C. Liu, A. Teke, M.A. Reshchikov, S. Doğan, V. Avrutin, S.J. Cho, J. Mater. Sci.: Mater. Electron. 18 (2007) 1119.
H. Morkoç, J. Appl. Phys. 98 (2003) 041301. [35] A.A. Dakhel, F.Z. Henari, Cryst. Res. Technol. 38 (2003) 979.
[7] K. Gurumurugan, D. Mangalaraj, S.K. Narayandass, C. Balasubramanian, Phys.
Status Solidi 143 (1994) 85.
[8] D.M. Carballeda-Galicia, R. Castanedo-Pérez, O. Jiménez-Sandoval, S. Jiménez-
Sandoval, G. Torres-Delgado, C.I. Zúñiga-Romero, Thin Solid Films 371 (2000) 105.