ARTICLE IN PRESS

Solar Energy Materials & Solar Cells 82 (2004) 315–330

Nano-structured CdTe, CdS and TiO2 for thin

film solar cell applications
R.S. Singha, V.K. Rangarib, S. Sanagapallia, V. Jayaramana,
S. Mahendraa, V.P. Singha,*
a
Center for Micro-Magnetic and Electronic Devices, Department of Electrical and Computer Engineering,
University of Kentucky, 453 F. Paul Anderson Tower, Lexington, KY 40506-0046, USA
b
Center for Advanced Materials, Tuskegee University, Tuskegee, AL, USA

Abstract

Nano-crystalline CdTe and nano-structured CdS films in crystalline, porous and fiber forms
were deposited on ITO-coated glass substrates. CdTe films exhibited a typical particle size of
10 nm and a blue shift in the absorption with an effective band gap of 2:8 eV: CdS films
exhibited a typical particle size of 15 nm and an effective band gap of 2:98 eV: Also, porous
CdS and porous TiO2 films were deposited on plastic substrates by a self-assembly method.
Typical pore sizes were 80 and 70 nm; respectively. These can be used in nano-structured solar
cell configuration where the pores are filled with a suitable absorber material. Additionally,
CdS fibers and nano-crystalline films of TiO2 on plastic substrates were fabricated. Typical
particle size in TiO2 was 10 nm and CdS fibers were about 80 nm wide and 1:5 mm long.
r 2004 Elsevier B.V. All rights reserved.

Keywords: CdS/CdTe solar cell; Nano-crystalline; Nano-porous; Thin film; TiO2

1. Introduction

Many nano-structured materials are now being investigated for their potential
applications in photovoltaic, electro-optical, micromechanical and sensor devices [1–
3]. Our interest lies in taking advantage of the benefits offered by nanotechnology to
make inexpensive and efficient solar cells on a large scale. To this end, nano-
structured layers in thin film solar cells offer three important advantages. First, due
to multiple reflections, the effective optical path for absorption is much larger than
the actual film thickness. Second, light generated electrons and holes need to travel

*Corresponding author. Tel.: +1-859-257-3092; fax: +1-859-257-3092.

E-mail address: vsingh@engr.uky.edu (V.P. Singh).

0927-0248/$ - see front matter r 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.solmat.2004.02.006
 ARTICLE IN PRESS
316 R.S. Singh et al. / Solar Energy Materials & Solar Cells 82 (2004) 315–330

over a much shorter path and thus recombination losses are greatly reduced. As a
result, the absorber layer thickness in nano-structured solar cells can be as thin as
150 nm instead of several micrometers in the traditional thin film solar cells [4].
Third, the energy band gap of various layers can be tailored to the desired design
value by varying the size of nano-particles. This allows for more design flexibility in
the absorber and window layers in the solar cell. In particular nano-structured CdS,
CdTe and TiO2 are of interest as window and absorber layers in thin film solar cells
[5–10]. Fabrication and characterization of these films constitute the focus of this
work.
Production of large-area solar cells would also require a process for inexpensive
fabrication of large periodic arrays of semiconductor nanostructures that will allow
for (a) controlled variations in the size and composition of the nanostructures, (b)
encapsulation of the semiconductor nanostructures in a rugged host material, (c)
flexibility to use a variety of substrate materials, and, preferably, (d) compatibility
with standard silicon fabrication techniques. Self-assembly is such a process. We
have used the self-assembly process to fabricate a variety of nano-structured films
including CdTe and CdS on ITO-coated glass substrates. In addition nano-porous
CdS and TiO2 films were fabricated on a plastic substrate with a view to making
devices on a lightweight, flexible substrate. Fabrication techniques and material
characteristics of these films and their applications to solar cells are described below.
Results on CdTe, CdS and TiO2 are presented in Sections 2, 3 and 4, respectively.

2. Nano-crystalline CdTe films

Nano-crystalline CdTe films are of interest because of their potential applications

as n-type window layers in a p–n homojunction thin film CdTe solar cell (Fig. 1) and
in electroluminescent displays devices [11–13]. CdTe nanocrystals were prepared by
microwave-assisted synthesis and films were cast from colloidal solutions containing

Graphite

Bulk CdTe
(Absorber layer, p-CdTe)

Nanocrystalline CdTe
(Window layer, n-CdTe)

ITO

Glass

Fig. 1. Device configuration of a Glass/ITO/n-Nano-CdTe/p-bulk CdTe/graphite solar cell.

 ARTICLE IN PRESS
R.S. Singh et al. / Solar Energy Materials & Solar Cells 82 (2004) 315–330 317

nano-CdTe. These films were characterized by optical absorption, photolumines-

cence spectroscopy, profilometer and scanning electron microscopy on glass and
plastic substrates for various conditions. The microwave-assisted synthesis was
carried out in a custom made 1000 W=2:45 GHz Sharp-R21-HT-microwave oven
supplied by Microwave Specialties Inc., USA. The oven’s cavity was modified to
include a complete refluxing system. In the present experiment, a 250 ml long neck
flask with a refluxing condenser assembly was used. All the reactions were carried
out with continuous heating at 50% power of the 1000 W; under flow of argon gas.
A coolant liquid ðB5 CÞ was circulated through the refluxing condenser throughout
the experiment. A few glass capillary tubes (3–5) were also added to the reaction
mixture to avoid the sudden liquid bumps.
1 g of cadmium (II) acetate dihydrate was dissolved in 100 ml of ethylene glycol
and then stoichiometric quantity ð0:40 gÞ of tellurium powder was added. The system
was purged for 10 min with argon and the reaction vessel was exposed to microwave
radiation continuously for 30 min: At this stage, the reaction mixture turned to
grayish color then the reaction mixture was diluted with ethanol to precipitate and
centrifuged at 5000 rpm: The precipitate was further washed with ethanol several
times and dried overnight under vacuum.
The nano-crystalline CdTe powder thus obtained by microwave assisted synthesis
was dispersed in methanol and coated on to ITO-coated glass. The resulting films
were examined with a FE-SEM and the electron micrograph shows a particle size of
about 10 nm (Fig. 2). The UV–VIS spectra recorded on this film shows a blue shift in

Fig. 2. FE-SEM image of a nanocrystalline CdTe film on ITO-coated glass substrate.

 ARTICLE IN PRESS
318 R.S. Singh et al. / Solar Energy Materials & Solar Cells 82 (2004) 315–330

Fig. 3. Absorption spectrum of a nanocrystalline CdTe film on ITO-coated glass substrate.

Fig. 4. Tauc’s law calculation for a nanocrystalline CdTe film on ITO-coated glass substrate.

absorption due to smaller particle size (Fig. 3). Tauc’s relation was used to determine
the effective band gap ðEeff Þ for the films. ðahnÞ2 was plotted against hn (Fig. 4) and
Eeff was obtained by extrapolation to zero value of ðahnÞ2 : Band gap estimate using
Tauc’s law indicates a value of about 2:8 eV as compared to a bulk CdTe band gap
of 1:5 eV: This increased band gap makes CdTe a good candidate for
electroluminescent display devices [11–13] and for window layers in n-CdTe/P-CdTe
homojunction solar cells (Fig. 1).

3. Nano-structured CdS films

Nano-structured CdS was fabricated in crystalline, porous and fiber forms. The
fabrication and characterization of these films are described below.
 ARTICLE IN PRESS
R.S. Singh et al. / Solar Energy Materials & Solar Cells 82 (2004) 315–330 319

3.1. Nano-crystalline CdS

CdS films were fabricated by three different techniques namely, solution growth
(type A), sonochemical (type B) and microwave-assisted synthesis (type C).
For types B and C, nano-crystalline CdS was first prepared in powder form by
sonochemical and microwave methods. SEM micrographs of the CdS powder
obtained by sonochemical and microwave synthesis methods showed spherical
particles, 15 nm in size. The X-ray diffraction pattern of the nano-crystalline CdS
powder shows peaks at 2y positions of 27 ; 44 and 52 corresponding to the (1 1 1),
(2 2 0) and (3 1 1) planes of the cubic CdS phase, respectively. The XRD patterns
obtained match with the JCPDS data (10-0454 Hawleyite). Sonochemical method
involves ultrasound irradiation of the sample during synthesis while microwave
method involves microwave irradiation. Sonochemistry is driven by acoustic
cavitation, formation, growth, and implosive collapse of bubbles in liquids irradiated
with high-intensity ultrasound. Sonochemical synthesis takes place when a strong
acoustic field is applied to an aqueous solution leading to cavity filled microbubbles.
Energy is transferred upon the growth and collapse of these microbubbles from the
macroscale acoustic wave to the microscale vapor inside the bubbles. Extremely high
pressures (hundreds of atmospheres) and temperatures (thousands of degree Kelvin)
result from the high localization [14–16].
Next, types B and C CdS films were deposited on ITO-coated glass by dip coating
in a suspension of the CdS powder. Type A films were deposited on ITO-coated glass
by the solution growth method. Further details on the synthesis of these films can be
found in our previous publication [5]. Particle size and the morphology of the films
were studied with SEM. Figs. 5(a)–(c) show the electron micrographs of CdS films
deposited by three different techniques. The solution growth films (Fig. 5(a)) have a
smooth morphology with a particle size of 110 nm and the film thickness is equal to
the particle size indicating a monolayer. Particle sizes in types B and C films, with a
typical value of 15 nm; are an order of magnitude smaller than in type A. A
hyperbolic band model [17] was used to estimate the energy gap as a function of
particle size.
g
/ 2 E ðp=rÞ2 =m 1=2 ;
E ¼ ½E 2 þ 2ðhÞ g ð1Þ
where Eg is the bulk band gap, r is the particle radius, and m is the electron effective
mass. The particle size corresponds to an energy gap of 2:98 eV for sonochemically
prepared CdS and shows a blue shift from the band gap of bulk CdS ð2:4 eVÞ:
X-ray diffraction measurements on the films indicated a cubic CdS phase. Optical
absorption plots of CdS films on ITO-coated glass were recorded using a UV–VIS
spectrophotometer. A comparison of optical absorption of CdS prepared by
sonochemical, microwave and solution growth methods is shown in Fig. 6. The onset
of absorption is at nearly 500 nm and a peak can be seen at close to 350 nm for type
B, while a peak at nearly 375 nm was observed for microwave (type C). Brus
describes this peak as 1S–1S transitions in CdS nanocrystals [18]. Thicker CdS films
show a second peak that can be attributed to a second excitation level [19].
Compared to bulk CdS deposited by solution growth, the optical absorption spectra
 ARTICLE IN PRESS
320 R.S. Singh et al. / Solar Energy Materials & Solar Cells 82 (2004) 315–330

Fig. 5. (a) FE-SEM image of CdS film prepared by solution growth method on ITO-coated glass
substrate. (b) FE-SEM image of CdS film prepared by sonochemical method on ITO-coated glass
substrate. (c) FE-SEM image of CdS film prepared by microwave method on ITO-coated glass substrate.

0.8

0.7

0.6
Absorption (Abs)

0.5

SG
0.4
MW
SC
0.3

0.2

0.1

0
300 350 400 450 500 550 600 650 700
Wavelength (nm)

Fig. 6. Comparison of UV–vis absorption spectra of CdS films fabricated by sonochemical, microwave
and solution growth methods on ITO-coated glass substrates.
 ARTICLE IN PRESS
R.S. Singh et al. / Solar Energy Materials & Solar Cells 82 (2004) 315–330 321

show a blue shift as a result of quantum confinement effects. Solution growth CdS
(type A) shows an optical absorption plot without a peak because of its larger
crystallite size.

3.2. CdS–Ag Schottky diodes

Schottky diodes were formed by depositing polycrystalline (bulk) CdS and nano
CdS on to ITO/glass substrates and subsequently depositing an Ag top contact. The
representative I–V curve of bulk CdS Schottky diode is shown (Fig. 7). The
characteristics of Schottky diodes on nano-CdS were substantially different from
those on bulk CdS (Fig. 8). In theory, nano-CdS/Ag Schottky diodes should have a
higher turn on voltage because of the larger band gap of nano-CdS, leading to
a higher barrier height. Yet the Schottky diodes fabricated on bulk CdS showed a
higher turn on voltage. We attribute this unexpected behavior to the fact that the
nanoCdS films are just 20 nm thick and the depletion layer may therefore extend into
ITO in this case. Since ITO is a degenerate semiconductor the effective depletion
layer width would be the thickness of nano-CdS film. Thus, in parts of the Schottky
barrier junction, the electrons would be able to ‘‘punch through’’, leading to a
reduction in the effective turn-on voltage. In comparison, the CdS thickness in
polycrystalline CdS layer is 110 nm:
Also, the I–V characteristics of the nano-CdS–Ag devices show some steps in the
plot. This can be understood as follows. As the nano-CdS is expected to be
completely depleted in this device the structure ITO/nano-CdS/Ag can be assumed
to behave like a metal/insulator/metal. The step behavior could be due to coulomb
blockade which is usually seen in such structures.

3.50E-06

3.00E-06 bulk CdS-Ag

2.50E-06

2.00E-06
Current (A)

1.50E-06

1.00E-06

5.00E-07

0.00E+00
-3 -2 -1 0 1 2 3
-5.00E-07
Voltage (V)

Fig. 7. Device configuration of a glass/ITO/n-nano-CdS/p-bulk CdTe/nano-CdTe/Au solar cell.

 ARTICLE IN PRESS
322 R.S. Singh et al. / Solar Energy Materials & Solar Cells 82 (2004) 315–330

1.50E-05

Nano CdS-Ag 1.30E-05

1.10E-05

9.00E-06

7.00E-06
Current (A)

5.00E-06

3.00E-06

1.00E-06

-2 -1 -1.00E-06 0 1 2
-3.00E-06

-5.00E-06
Voltage (V)

Fig. 8. I–V curve of a glass/ITO-bulk CdS–Ag Schottky diode.

Au/Al

Nano CdTe

Bulk CdTe
(Absorber layer, p-CdTe)

Nanocrystalline CdS
(Window layer, n-CdS)

ITO

Glass

Fig. 9. I–V curve of a glass/ITO-nano CdS–Ag Schottky diode.

An application of nano-structured CdS to solar cells is illustrated in Fig. 9. Wider

band gap makes it a more effective window layer. We are now in the process of using
these larger band gap CdS films as window layers in the fabrication of experimental
CdS–CdTe thin film solar cells.

3.3. Porous CdS films

Our objective is to make porous structure of CdS, deposit an absorber material

like, copper indium diselenide, copper sulfide or cadmium telluride into the pores
 ARTICLE IN PRESS
R.S. Singh et al. / Solar Energy Materials & Solar Cells 82 (2004) 315–330 323

and study the nano-structured, CdS-based heterojunction solar cells. This structure
will allow for a device design involving an n-type material as a porous film and an
extremely thin p-type absorber material filling the pores. The advantages are that the
transport path for excited charge carriers in the absorber is reduced and at the same
time, the optical path for photon absorption is increased. Due to the nano-porous
structure of the cell, the distance that photoexcited electrons must travel within their
lifetime can be reduced to less than 100 nm [20–22]. This design is also of critical
importance in case of organic solar cells and hybrid junctions between inorganic and
organic layers, where the traditional non-nano-structured designs are not beneficial
for electron/hole transport. This is due to the fact that organic films tend to have
short diffusion lengths and because of anisotropy, the conducting axis in the organic
layer is not always along the film thickness.
Nano-porous CdS films were fabricated on a plastic substrate by using ultrasound
irradiation during dip-coating. We term this process as ultrasound-assisted synthesis
(UAS). Ultrasound irradiation was used as a tool to fabricate porous semiconductor
films; it is known to cause significant effect on processes that occur at the surface of
the solids immersed in liquids. Ultrasonic irradiation generates acoustic cavitation in
the liquid, resulting in the formation of microbubbles. These microbubbles collapse
on colliding with the solid surface causing erosion [23]. We believe that this process is
responsible for the formation of a porous structure in our CdS films. In our method,
the plastic substrate is dipped in CdS solution and exposed to ultrasound irradiation;
it is thought that the microbubbles collapse at the surface of substrate and cause
erosion of the CdS particles stuck to the substrate. Furthermore, since ultrasonicator
generates acoustic cavitation uniformly throughout the liquid, the pores formed are
uniform throughout the film.
SEM results showed uniform pore structure with a pore size of about 80 nm
(Fig. 10). In contrast, the films deposited on ITO/glass substrates with UAS showed
a crystalline film with voids. The optical absorption results of all these films showed a
blue shift and an increase in the effective band gap (Figs. 11 and 12).
The method reported above is applicable to other materials and constitutes a
practical route to prepare inexpensively, large area, flexible, lightweight, porous
metal sulfide semiconductor films with an ordered nano-porous array structure.
These films can serve as the underlying layers not only for the advantageous
inorganic and organic solar cells designs as explained above, but also for sensors and
other optoelectronic devices. This is because the large surface area in the nano-
porous films can provide a highly active reaction interface and enhanced mass
transfer properties.

3.4. CdS fibers

We fabricated nano-fibers of CdS by a combination of solution chemistry and

ultrasound. High-intensity, high-frequency ultrasound irradiation was found to play
a critical role in the crystallization and the preferential one-dimensional growth of
the nano-wires of CdS in the solution. During ultrasound irradiation of liquid-
powder slurries, cavitations and shock waves it creates can accelerate solid particles
 ARTICLE IN PRESS
324 R.S. Singh et al. / Solar Energy Materials & Solar Cells 82 (2004) 315–330

Fig. 10. FE-SEM image of porous CdS film on plastic substrate.

0.2

0.175

0.15 SG
Absorption (a.u)

0.125 SC
MW
0.1

0.075

0.05

0.025

0
300 350 400 450 500 550 600 650 700
Wavelength (nm)

Fig. 11. Comparison of UV–vis absorption spectra of porous CdS films fabricated by sonochemical,
microwave and solution growth methods on plastic substrates.

to high velocities. The inter-particle collisions that result are capable of inducing
striking changes in morphology, composition, and reactivity of the solids. During
the inter-particle collisions, the particles can be driven together at sufficiently high
speeds to induce effective melting at the point of collision. It is thought that the
energy generated during collision induced the crystallization and 1-D growth of CdS
 ARTICLE IN PRESS
R.S. Singh et al. / Solar Energy Materials & Solar Cells 82 (2004) 315–330 325

0.9

0.8

0.7
Absorption (Abs.)
0.6

0.5 SG:Porous CdS

0.4

0.3

0.2

0.1

0
300 400 500 600 700
Wavelength (nm)

Fig. 12. Optical absorption of porous CdS films by solution growth method.

nanowires. It was observed that temperature, pressure, and reaction time played
important roles in the solvothermal fabrication of nano-scale-sized CdS fibers. Only
when the reaction was carried out in sealed systems, in the presence of high intensity
and high frequency ultrasound, the CdS nano-fibers were formed. The physical
effects of ultrasound cavitations in liquid–solid systems are primarily responsible for
the enhancement of the generation of surface damage at the liquid–solid interface by
shock waves or microjets and fragmentation of friable solids to increase the surface
area. The impingement of microjets or shock waves on the solid surface creates the
localized erosion, which is responsible for the ultrasound cleaning and many other
sonochemical effects on heterogeneous reactions [24]. This phenomenon was also
observed in the case of porous CdS films fabricated using an ultrasound-assisted
synthesis, as described in the previous section.
The typical length of the fibers was about 1:5 mm and the diameter was about
80 nm: The fiber length was found to be a function of reaction time. The typical
scanning electron micrograph of CdS nano-fiber is shown in Fig. 13. The optical
absorption measurements of CdS fibers dispersed on an ITO/glass substrate indicate
a broad hump around 475 nm and a peak at 330 nm (Fig. 14). These fibers are
potential candidates for channels in thin film transistors and have applications in
flexible, wearable and disposable electronics.

4. Nano-structured TiO2 films

TiO2 films were deposited on plastic substrates by the sol–gel method. The TiO2
sol was prepared using the standard procedure starting with high-purity reagents,
titanium tetra isopropoxide, isopropanol and nitric acid (70% redistilled). This
method is based on the hydrolysis of metal alkoxides or ethoxides in alcoholic
solutions in the presence of acid catalysts. The procedure for the preparation
involved the dissolution of TTIP as precursor in isopropanol as a solvent, followed
 ARTICLE IN PRESS
326 R.S. Singh et al. / Solar Energy Materials & Solar Cells 82 (2004) 315–330

Fig. 13. FE-SEM image of CdS fibers on ITO-coated glass substrate.

Optical Absorption of CdS Fibers

0.32

0.3

CdS fibers
0.28
Absorption (a.u)

0.26

0.24

0.22

0.2
300 350 400 450 500 550 600 650 700
Wavelength (nm)

Fig. 14. Optical absorption of CdS fibers on ITO-coated glass substrate.

 ARTICLE IN PRESS
R.S. Singh et al. / Solar Energy Materials & Solar Cells 82 (2004) 315–330 327

by adding deionized water and nitric acid (70%). The reaction was carried out with
nitrogen gas flowing into the hood. TTIP was added to isopropanol and to this
solution, deionized water was added dropwise while stirring. Nitric acid was added
dropwise slowly to TTIP, isopropanol and deionized water while still stirring under
the hood. The stirring was continued for 2 h and then the solution was covered with
a parafilm and placed under the hood. In a typical preparation of 0:1 M TiO2 sol,
1 ml of TTIP, 0:05 ml of HNO3 (70% distilled), 0:1 ml of deionized water and
32:7 ml of isopropanol were used. The films were deposited by means of spin or dip
coating (withdrawal rate 0:5 mm=s) on a clean substrate. Before coating, the
substrates were cleaned with a flow of acetone and deionized water with intermediate
drying in flowing nitrogen. After cleaning, substrates were either spin or dip coated.
For obtaining the nano-porous films, subsequent to the coating process, the
substrates were dried under hot water vapor; nitrogen was passed through a hot
bubbler, with the humidity controlled by the flow rate of the nitrogen gas
(E1:3 cm=s through the chamber). The humidity and temperature of the chamber
were monitored with a digital hygrometer. The nano-crystalline TiO2 films, on the
other hand, were spin or dip coated from a TiO2 sol ð0:01 MÞ on to the plastic
substrates and dried in ambient.
The material characterization of the films was performed using a Hitachi S-900
FE-SEM for studying morphology and pore structure. A Shimadzu UV-2501 UV–
Vis spectrometer was used for optical absorption studies. The thickness of the films
was measured using an Alpha step profilometer. The scanning electron micrographs
(Figs. 15(a) and (b)) show a porous structure over large areas. The pore size was
estimated to be 70 nm: The walls of the pores were found to be thicker than in case of
films made via Benard-marangoni convection [25]. The scanning electron micro-
graph of a typical nano-crystalline TiO2 film fabricated by drying the films in
ambient is shown in Fig. 16. These films are smooth and have a particle size of about
10 nm: The optical absorption of the films was measured in the near UV and visible

Fig. 15. (a) FE-SEM image of porous TiO2 on plastic substrate, at low magnification. (b) FE-SEM image
of porous TiO2 on plastic substrate, at high magnification.
 ARTICLE IN PRESS
328 R.S. Singh et al. / Solar Energy Materials & Solar Cells 82 (2004) 315–330

Fig. 16. FE-SEM image of Crystalline TiO2 on plastic substrate.

range using a Shimadzu spectrometer. The spectral dependence, in general, shows

one region of fundamental absorption edge ðEg Þ and the other region characterized
by interference peaks at higher wavelengths. All of the three nano-porous films of
different thicknesses showed a peak at 334 nm in the optical absorption spectra, but,
as expected, the magnitude of absorption is higher in thicker films (Fig. 17).
Just as in case of porous CdS films, porous TiO2 films can serve as the underlying
layers for efficient solar cell designs. Absorber materials like copper indium
diselenide, copper sulfide or cadmium telluride can be deposited into the pores of
TiO2 and nano-structured, heterojunction solar cells obtained. Also, the large
surface area in the nano-porous films can provide a highly active reaction interface
and enhanced mass transfer properties.

5. Conclusions

Nano-crystalline CdTe films were deposited on ITO coated glass substrates and
characterized. The results indicated particle sizes of 10 nm and a blue shift in the
 ARTICLE IN PRESS
R.S. Singh et al. / Solar Energy Materials & Solar Cells 82 (2004) 315–330 329

0.5

0.45

0.4

0.35
Absorption (a.u)

0.3 NPTiO2-TP1
NPTiO2-TP2
0.25 NPTiO2-TP3
0.2 TiO2-TP5
TiO2-TP6
0.15

0.1

0.05

0
300 350 400 450 500 550 600 650 700
Wavelength (nm)

Fig. 17. Comparative UV–vis absorption spectra of porous and crystalline TiO2 films on plastic
substrates.

absorption with an effective band gap of 2:8 eV: This opens the possibility of using
nano-crystalline n-type CdTe as a window layer in an n-CdTe/p-CdTe homojunction
solar cell. Nano-crystalline CdS films on ITO-coated glass substrates exhibited
particle sizes of 15 nm and an effective band gap of 2:98 eV as compared to the
2:4 eV value for the band gap of bulk CdS. This makes nano-crystalline CdS a better
window material in an n-CdS/p-CdTe heterojunction solar cell. We are now in the
process of evaluating these advantages by fabricating test structures of n-CdS/p-
CdTe and n-CdTe/p-CdTe solar cells.
Porous CdS and porous TiO2 films were deposited on plastic substrates by a self-
assembly method. Typical pore sizes were 80 and 70 nm; respectively. These can be
used in nano-structured solar cell configuration where the pores are filled with a p-
type absorber material. Due to the nano-structured character of the absorber, the
transport path for the light generated electrons in the absorber is reduced. At the
same time, the optical path for photon absorption is increased due to multiple
reflections.
Also, CdS fibers and nano-crystalline films of TiO2 on plastic substrates were
fabricated. Typical particle size in nano-crystalline TiO2 was 10 nm and CdS fibers
were about 80 nm wide and 1:5 mm long.

Acknowledgements

The authors thank the Kentucky Science and Engineering foundation for support
through Grants KSEF-148-502-02-27 and KSEF-148-502-03-68. One of the authors
 ARTICLE IN PRESS
330 R.S. Singh et al. / Solar Energy Materials & Solar Cells 82 (2004) 315–330

(VKR) thanks Dr. Shaik Jeelani, Director, T-CAM, Tuskegee University for support
and encouragement.

References

[1] G.M. Whitesides, Bartosz Grzybowski, Science 295 (2002) 2418–2421.

[2] Xiangfeng Duan, Chunming Niu, Vijendra Sahi, Jian Chen, J.W. Parce, S. Empedocles, J. Goldman,
Nature 425 (2003) 274–278.
[3] M.C. McAlpine, R.S. Friedman, Song Jin, Keng-hui Lin, Wayne U. Wang, C.M. Lieber, Nano Lett.
3 (2003) 1531–1535.
[4] K. Ernst, A. Belaidi, R. Konenkamp, Semicond. Sci. Technol. 18 (2003) 475.
[5] V.P. Singh, R.S. Singh, G.W. Thompson, V. Jayaraman, S. Sanagapalli, V.K. Rangari, Sol. Energy
Mater. Sol. Cells 81 (2004) 293–303.
[6] V.P. Singh, J.C. McClure, Sol. Energy Mater. Sol. Cells 76 (2003) 369–385.
[7] V.P. Singh, O.M. Erickson, J.N. Chao, J. Appl. Phys. 78 (7) (1995) 4538–4542.
[8] D.L. Linam, V.P. Singh, J.C. McClure, G.B. Lush, X. Mathew, P.J. Sebastian, Sol. Energy Mater.
Sol. Cells 70 (2001) 335–344.
[9] M.E. Calixto, J.C. McClure, V.P. Singh, P.J. Sebastian, Sol. Energy Mater. Sol. Cells 63 (4) (2000)
325–334.
[10] X. Mathew, J.P. Enriquez, P.J. Sebastian, J.C. McClure, V.P. Singh, Sol. Energy Mater. Sol. Cells 63
(4) (2000) 355–365.
[11] Wei Chen, D. Grouquist, J. Roark, J. Nanosci. Nanotechnol. 2 (2002) 47.
[12] V.P. Singh, A. Aguilera, A. Garcia, D.C. Morton, IEEE Trans. Electron Dev. 48 (10) (2001)
2242–2248.
[13] Mingyuan Gao, C. Lesser, S. Kirstein, H. Mohwald, A.L. Rogach, H. Weller, J. Appl. Phys. 87
(2000) 2297.
[14] Cuiling Gong, D.P. Hart, J. Acoust. Soc. Am. 104 (1998) 2675–2682.
[15] K.S. Suslick, Sonochemistry: a physical perspective, AIP Conference Proceedings, Nonlinear
Acoustics at the Turn of the Millennium, Vol. 524, 2000, pp. 95–104.
[16] K.S. Suslick, Sonochemistry, Science 247 (1990) 1439–1445.
[17] J.A. Akintunde, Phys. Status Solidi. A 179 (2000) 363.
[18] L. Brus, J. Chem. Phys. 90 (1986) 2555.
[19] Y. Wada, H. Kuramoto, J. Anand, T. Kitamura, T. Sakata, H. Mori, S. Yanagida, J. Mater. Chem.
11 (2001) 1936.
[20] Yu, T. Isobe, M. Senna, Mater. Res. Bull. 30 (1995) 975.
[21] D.J. Lockwood, A. Wang, B. Bryskiewicz, Solid State Commun. 89 (1994) 587.
[22] K. Ernst, A. Belaidi, R. Konenkamp, Semicond. Sci. Technol. 18 (2003) 475.
[23] A. Vogel, W. Lauterborn, R. Timm, J. Fluid Mech. 206 (1989) 299.
[24] K.S. Suslick (Ed.), Ultrasound: Its Chemical, Physical and Biological Effects, VCH, Weinheim, 1988.
[25] R.S. Singh, C.A. Grimes, E.C. Dickey, Mater. Res. Innov. 5 (2002) 178.