Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Download as pdf or txt
Download as pdf or txt
You are on page 1of 32

Solar Energy Materials and Solar Cells 52 (1998) 313—344

Semiconductor thin films by chemical bath


deposition for solar energy related applications
P.K. Nair*, M.T.S. Nair, V.M. Garcı́a, O.L. Arenas, Y. Pen8 a,
A. Castillo, I.T. Ayala, O. Gomezdaza, A. Sánchez, J. Campos,
H. Hu, R. Suárez, M.E. Rincón
Department of Solar Energy Materials, Centro de Investigacio& n en Energı& a, Universidad Nacional Autonoma de
Me& xico, Temixco, Morelos 62580, Mexico
Received 1 October 1996; received in revised form 20 May 1997

Abstract

In this paper we present the basic concepts underlying the chemical bath deposition
technique and the recipes developed in our laboratory during the past ten years for the
deposition of good-quality thin films of CdS, CdSe, ZnS, ZnSe, PbS, SnS, Bi S , Bi Se , Sb S ,
2 3 2 3 2 3
CuS, CuSe, etc. Typical growth curves, and optical and electrical properties of these films are
presented. The effect of annealing the films in air on their structure and composition and on the
electrical properties is notable: CdS and ZnS films become conductive through a partial conver-
sion to oxide phase; CdSe becomes photosensitive, SnS converts to SnO , etc. The use of
2
precipitates formed during deposition for screen printing and sintering, in polymer composites
and as a source for vapor-phase deposition is presented. Some examples of the application
of the films in solar energy related work are presented. ( 1998 Elsevier Science B.V. All
rights reserved.

Keywords: Thin films; Growth curves; Optical and electrical properties

1. Introduction

Chemical bath deposition technique is well suited for producing large-area thin
films for solar energy related applications. In this paper we would share our

* Corresponding author.

0927-0248/98/$19.00 ( 1998 Elsevier Science B.V. All rights reserved


PII S 0 9 2 7 - 0 2 4 8 ( 9 7 ) 0 0 2 3 7 - 7
314 P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344

experience in this technique, acquired at the Laboratorio de Energia Solar (presently


Centro de Investigación en Energı́a) during the past 10 yr, with researchers in this field
and others who would be interested to initiate work in this research area.
Chemical bath deposition is a technique in which thin semiconductor films are
deposited on substrates immersed in dilute solutions containing metal ions and
a source of hydroxide, sulfide or selenide ions. The earliest reported work dates back
to 1919 and dealt with the deposition of PbS thin films [1]. The basic principles
underlying the chemical bath deposition of semiconductor thin films and early
research work in this area have been presented in a 1982 review article [2], which has
inspired many researchers to initiate work in this area. The subsequent progress in
this area is contained in a 1991 review article [3], which lists literature on more than
35 compounds prepared by this technique and related references. Recipes for the
chemical bath deposition of a number of such compounds are given in a recent paper
[4]. The number of possible materials to be produced through the technique is bound
to multiply in subsequent years. This is due to the feasibility of producing multilayer
films by this technique — the annealing of which promotes interfacial diffusion of metal
ions and the production of new materials with improved thermal stabilities. For
example, interfacial diffusions in chemically deposited PbS—CuS and ZnS—CuS coat-
ings result in materials such as Pb Cu S and Zn Cu S with p-type conductivities and
x y z x y z
are stable up to a temperature of 300°C [5]. Annealing of Bi S (bismuthinite)—CuS
2 3
(covellite) coating at temperatures of 250—300°C leads to the formation of a new
compound, Cu BiS (wittichenite), with p-type conductivity [6].
3 3
Among the first applications of chemically deposited semiconductor thin films, we
find PbS and PbSe photodetectors [7]. Even though chemically deposited CdS films
were developed during 1960s, the photodetector applications of CdS were confined to
screen-printed and sintered layers [8]. However, chemically deposited CdSe thin films
were found appropriate for photodetector applications [9]. In the late 1970s and early
1980s the motivation in the work on chemically deposited thin films has been their
prospective solar energy applications. One of the earlier developments toward this
was in solar absorber coatings [10], followed by specific application of such coatings
in evacuated all-glass tubular collectors [11]. Application of the chemically deposited
films in solar control coatings was suggested in 1989 [12]. Subsequently the perfor-
mance evaluation of such coatings was reported [13].
The use of chemically deposited semiconductor films to develop photo-
electrochemical solar cells was of basic interest, most importantly, in the case of CdS
and CdSe thin films [2,14,15]. Recent work on chemically deposited CdSe and Sb S
2 3
films with an incorporated WO component has shown improved conversion efficien-
3
cies and stabilities in photoelectrochemical solar cell configuration [16].
The integration of chemically deposited semiconductor thin films in thin-film solar
cell has only a short history. In 1990, a thin layer of chemically deposited CdS thin film
was integrated into a structure Mo—CuInSe —CdS—ZnO producing approximately
2
11% of conversion efficiency [17]. This was followed by improved cell design resulting
in record efficiencies of '17% [18], in which the chemically deposited CdS film of
&0.05 lm continued to play the role of an inevitable component. The incorporation
of highly resistive CdS film in the solar cell structure, (p)CuInSe —CdS—(n)CdS, was
2
P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344 315

identified in 1979—80 as an essential step towards improving the solar cell stability
[19]. Theoretical calculations done in 1982 [20] predicted that the thickness of the
CdS film should be as small as possible to increase the cell efficiency. The chemical
deposition technique, which excels in providing complete surface coverage at very
small film thicknesses was therefore chosen to produce this very thin film with
excellent success [21].
The use of chemically deposited n-CdSe or n-Sb S films with incorporated WO
2 3 3
phase as an active absorber material in Schottky barrier solar cells has been proven
factible. For example, » of 0.72 V; I of 14.1 mA cm~2; FF of 0.7 and conversion
0# 4#
efficiency of 5.5% have been demonstrated in ITO-(n)CdSe(5 lm)-Pt/Ni/Au(130 A_ )
Schottky cells [16,22,23]. Similarly, the fabrication of heterojunction solar cells using
chemically deposited Sb S thin films on (p)Si/(p)Ge/(p)InP wafers has been demon-
2 3
strated [16,22,24].
The photoaccelerated chemical deposition on PbS thin films was reported in 1991
[25] in which the illuminated region in a growing thin film showed accelerated
deposition. The application of this effect in producing images in PbS thin film [26]
and Bi S thin film [27] has been reported. The potential use of the photoacceleration
2 3
effect in attaining high deposition rate is evident.
The above range of prospective applications, particularly, in the area of solar
energy conversion and energy efficiency has prompted renewed interest in chemically
deposited semiconductor thin films. The technique is ideally suited for the production
of large area thin films as required for solar energy applications. In this paper we shall
outline the technique for the preparation of a number of chemically deposited
semiconductor thin films. The importance of post-deposition treatment in the modifi-
cation of chemically deposited thin films for various applications and the prospects for
producing new solar energy materials through interfacial diffusion of metal ions will
be discussed. The use of the precipitate produced in the deposition bath during thin-
film deposition, for the production of semiconductor coatings by other deposition
techniques will also be presented.

2. Experimental

2.1. Chemical bath deposition

The basic advantages of chemical bath deposition have been highlighted by various
authors [1—4,16]. The technique is applicable for the deposition of highly insoluble
compounds. For example, the solubility of CuS at room temperature in water is
estimated to be 10~36 g dm~3, or in other words, the product of the ionic concentra-
tions of the Cu2` and the S2~ ions in the saturated aqueous solution of CuS is
10~36 mol2dm~6. This is the solubility product (SP). If in an arbitrarily constituted
aqueous solution of Cu2` and S2~ ions, the product of their ionic concentration (IP)
is '10~36 mol2 dm~6, the excess ions will be precipitated as CuS. The details of
these concepts are given in standard sources [28]. A list of the solubility products of
sulfides and hydroxides of various metals is given in Ref. [29].
316 P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344

In the above-mentioned event, if the precipitation is controlled through the use of


suitable complexing agents such as triethanolamine for the metal ions (so as to reduce
the amount of free metal ions [30]) and the amount of sulfide, selenide, or hydroxo
ions in the bath is controlled through setting up of appropriate chemical equilibria
[2,3], thin-film deposition can take place. A further condition to be satisfied is the
availability of nucleation centers over the substrate. Such centers are normally formed
through the adsorption of metal hydroxo species over the surface. The hydroxo group
would be substituted by the sulfide or selenide ions which would thereby form an
initial layer of the metal chalcogenide [31]. The deposition of the thin film takes place
through the condensation of the metal and sulfide/selenide ions on this initial layer,
which acts as a catalytic surface.
The following compositions for the chemical deposition baths have been developed
in our laboratory over many years. They have been tested for their reproducibility and
have been found to be very reliable for the deposition of thin films in the thickness
range 0.05—0.5 lm. The given volumes of the constituents are the quantities required
to prepare 100 ml of the deposition bath. The usual substrates are microscope glass
slides, 75 mm]25 mm]1mm, cleaned in a detergent solution and rinsed well in
running tap water. These are mounted nearly vertical on the wall of a 100 ml beaker
containing the bath. Up to six substrates may be placed in a 100 ml beaker in this way.
Deposition on a large-area commercial sheet glass may be made by slowly laying
them down on a shallow tray containing the deposition bath. The sheet glass would be
separated from the bottom of the tray by spacers in such a way that an approximately
5 mm thick layer of deposition bath separates the bottom surface of the sheet glass
and the bottom of the tray. The thin-film deposition would take place at the bottom
surface of the sheet glass. Coatings on sheet glass of 60 cm] 60 cm are routinely made
in our laboratory by this technique using approximately 2000 ml each of the depos-
ition bath. Relatively low-cost reagents, of purity not more than 98%, are usually
sufficient for the preparation of the deposition baths.
CdS thin film [32]: 30 ml of 0.1 M Cd(CH COO) ) 2H O, 8—12 ml of 1 M sodium
3 2 2
citrate, 15 ml of 1.5 M NH OH, 5—10 ml of 1 M thiourea, and the rest deionized water
4
to make the volume up to 100 ml. The deposition is allowed to proceed at 50—70°C for
up to 12 h. Very smooth and uniform coatings of CdS are formed except at long
durations of deposition.
CdSe thin films [33,34]: 30 ml of 0.1 M Cd(NO ) ) 4H O, 12 ml of 1 M sodium
32 2
citrate, 1.2 ml of 30% (&15 M) NH OH, 0.4 g of N,N-dimethylselenourea dissolved
4
in 30 ml of freshly prepared 0.01 M Na SO , and the volume made up to 100 ml with
2 3
water. Depositions may be made at room temperature (24 h) or at temperatures up to
60°C (8 h); at prolonged deposition there is risk of the film peeling from the glass
substrates.
ZnS thin films [35]: 5 ml of 1 M zinc sulfate, 4.4 ml of NH /NH Cl (pH 10), 5.4 ml of
3 4
50% triethanolamine, 2 ml of 1 M thioacetamide, and the rest deionized water to
make up to 100 ml by volume. Depositions are made at 50°C for about 6 h or at room
temperature for about 20 h to obtain ZnS films of 0.2 lm thickness.
ZnSe thin films [36]: 35 ml of 0.1 M zinc acetate, 16 ml of 0.8 M sodium citrate, 5 ml
of 7.4 M ammonium hydroxide, and 20 ml of 0.07 M N,N-dimethylselenourea, and
P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344 317

the volume made up to 100 ml with deionized water. Depositions may be made at
room temperature or in an oven at temperatures up to 60°C.
SnS thin films [37]: 1 g SnCl ) 2H O dissolved in 5 ml of acetone, 12 ml of 50%
2 2
triethanolamine, 8 ml of 1 M thioacetamide, 10 ml of 4 M NH OH and the rest water
4
to make up the volume to 100 ml. The depositions may be made at temperatures up to
80°C to obtain thin films of thickness 0.7 lm.
PbS thin films [38]: 5 ml of 1 M Pb(CH COO) ) 3H O, 20 ml of 1 M NaOH, 5 ml
3 2 2
of 1 M thiourea, 5 ml of 1 M triethanolamine, and the rest water. The quality of films
is best when depositions are done at room temperature.
Bi S thin films [39,40]: 10 ml of K0.5 M Bi(NO ) ) 5H O solution, 8 ml of 50%
2 3 33 2
triethanol-amine, 4 ml of 1 M thioacetamide, and the rest water. The depositions are
made at room temperature; there is risk of peeling of the film from the substrate at
intermediate durations (5—7 h), but are seen to be adherent when the films are allowed
to remain in the bath for more than 24 h. Improved adhesion of the films are possible
when the films are deposited over a substrate layer of ZnS [35].
Bi Se thin films [41]: 7 ml of 0.5 M Bi(NO ) solution, 7 ml of 50% tri-
2 3 33
ethanolamine, 20 ml of 0.07 M N,N-dimethylselenourea solution prepared freshly
in 0.01 M Na SO , and the rest water. Films of 0.2 lm may be deposited in
2 3
about 9 h at 40°C or in about 60 h at room temperature. The films adhere well to glass
substrates.
CuS thin films [42]: 5 ml of 1 M CuCl .3H O, 4 ml triethanolamine, 8 ml of 30%
2 2
NH OH, 10 ml of K1 M NaOH, 6 ml of 1 M thiourea, and the rest water. Depos-
4
itions are made at room temperature for best results; the films peel from glass
substrates during prolonged deposition, but it may be prevented by using a ZnS
substrate film [35].
Cu Se thin films: 10 ml of 0.5 M CuSO ) 5H O solution, 1.5 ml of 30%
2~x 4 2
NH OH, 12 ml of K0.4 M Na SeSO solution (prepared by reflexing 4 g of selenium
4 2 3
powder in 100 ml of 1 M Na SO solution for about 3 h), and the rest water. At room
2 3
temperature, films of approx. 0.2 lm thickness are deposited in 7 h. The value of
x may depend on the exact bath composition and may even gradually change with the
duration of deposition.
Sb S thin films: 0.65 g SbCl dissolved in 2.5 ml of acetone, 25 ml of 1 M sodium
2 3 3
thiosulfate, and the rest water to make up the volume to 100 ml. The depositions were
done at 10°C in a refrigerator.
The compositions of deposition baths for many other compound semiconductor
thin films developed in other laboratories are summarized in Ref. [3].

2.2. Characterization

The X-ray diffraction patterns of the films were recorded on a Seimens D500
machine. The optical transmittance and the near-normal specular reflectance spectra
of the samples were recorded on a Shimadzu 3101 PC UV VIS NIR spectro-
photometer. For the electrical measurements, a pair of silver paint electrodes of 5 mm
length at 5 mm separation were printed on the film surface to serve as contacts. The
measurements were done using a computerized data acquisition system.
318 P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344

3. Results and discussion

3.1. Growth of thin films

In general, the growth of good-quality semiconductor thin films by the chemical


bath deposition technique proceeds at a slow pace. The technique is ideally suited for
producing uniform films with thickness in the 0.05—0.3 lm range in most cases. Film
thickness up to a micron is possible in some cases. Thickness up to a few micron in
chemically deposited CdSe and Sb S thin films have been reported from other
2 3
laboratories [16]. Higher deposition rates and higher film thicknesses are usually
accompanied by powdery deposits and a lack of specular reflection. Fig. 1 shows a few
examples of the films deposited in our laboratory. All the films shown in the figure are
specularly reflective. The bath compositions are stated in Section 2.1.
Fig. 2 illustrates the quality of the films: the transmittance, ¹(j)%, and the specular
reflectance, R(k)%, add up to nearly 100 at wavelengths longer than the threshold for
optical absorption across the band gap. Arrows in the figure indicate the values listed
for the optical band gap 1.03 eV (1204 nm) and 1.88 eV (659.5 nm) for Bi Se and
2 3
Sb S , respectively [41,43].
2 3

Fig. 1. Film thickness versus duration of deposition for different semiconductor thin films obtained by the
chemical bath deposition technique. The bath temperature is indicated in parentheses; the bath composi-
tion is given in the text in Section 2.1.
P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344 319

Fig. 2. The near-normal reflectance (R%) and transmittance (T%) curves of Sb S and Bi Se thin films
2 3 2 3
indicating the presence of interference peaks. The sum of R(%)#¹(%) is nearly 100% at long wavelengths,
indicating that the films possess nearly smooth surfaces. Arrows indicate the positions of absorption edges
reported in Refs. [41,43].

3.2. Quantum confinement effect

The effect of crystallite size on the optical band gap of semiconductor thin films
has been discussed by many researchers [44]. It is known that for an electron
in a quantum well of width ¸, the quantized energy levels E are given by [45]
n
E "1(h/2p)2p2n2(m )~1/¸2. The ground-state energy is given by E "1(h/2p)2
n 2 % 1 2
p2(m )~1/¸2, where h is the Planck’s constant and m is the effective mass of electron.
% %
In the case of a polycrystalline semiconductor with crystalline grain width ¸,
this leads to an enhancement of bandgap, (E ) as compared to that of a bulk
' 1#
320 P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344

semiconductor crystal [46]:

(E ) "(E ) #1(h/2p)2p2(m~1#m~1)/¸2!(1.8/e¸),
' 1# ' "6-, 2 % )
where m is the effective mass for hole and the last term represents the screening term
)
in a medium of permittivity e. (E ) approaches (E ) for ¸'50 nm, but takes
' 1# ' "6-,
values considerably higher than the bulk value for ¸(5 nm. For example, in
chemically deposited CdSe films with grain sizes (5 nm, optical band gap of up to
0.7 eV higher than the bulk value (of 1.74 eV) has been reported [47].
Fig. 3 illustrates this effect in the case of Sb S thin films chemically deposited at
2 3
10°C. The plots of square of the optical absorption coefficient (a) versus the photon
energy (hl) indicate direct band gaps for this film of about 0.7 eV higher than the bulk
value (1.88 eV) for the band gap [43]. The grain size of these films are less than 5 nm;
the X-ray diffraction patterns did not indicate any diffraction peak. These results for
Sb S films are in agreement with those reported by other workers [48].
2 3

3.3. Preferential orientation of the films

Preferential orientation of the crystallites perpendicular to the plane of the sub-


strate is a common feature observed in the case of chemically deposited CdS and CdSe

Fig. 3. The very small crystallites in chemically deposited thin films usually give rise to variation in band
gaps due to quantum confinement effects; the case of Sb S films of 50, 90, and 120 nm thickness is illustrated.
2 3
P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344 321

Fig. 4. Preferential orientation of the crystalline grain, with the c-axis perpendicular to the plane of the
substrates is observed in the case of chemically deposited thin films in the thickness range 110—920 nm.

thin films. Fig. 4 shows the typical case of CdS thin films deposited at 60°C, using
citrate as a complexing agent as described in Section 2.1. The orientation is so
pronounced that only the diffraction peak corresponding to the (0 0 2) planes is
noticeable. The figure also shows that at very small film thicknesses the films appear
to be amorphous or possess very small crystallites.

3.4. Photocurrent response

Fig. 5 shows the photocurrent response of a number of chemically deposited thin


films. Except for CdS, the films do not exhibit large changes in the current upon
illumination. Most films are highly resistive in the dark; the electrical conductivities in
the dark range from 10~8 to 10~6 )~1 cm~1. Exceptions are CuS and Cu Se thin
2~x
films: the film conductivities are in the 103 to 104 )~1 cm~1 range. Such high
electrical conductivities produce free carrier absorption in the infrared region, very
small optical transmittance and high reflectance in the infrared region. These features
in the optical transmittance and reflectance spectra are shown in Fig. 6. The high
electrical conductivity arises from copper deficiency in the films, which make them
p-type [49]. This is a feature of copper chalcogenide films prepared by vacuum-based
322 P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344

Fig. 5. Photocurrent response of various semiconductor thin films prepared in our laboratory by chemical
bath deposition technique.
P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344 323

Fig. 7. Variation in the photosensitivity of the films produced by annealing of a CdSe film (&150 nm) in air
for 1 h each at different temperatures.

techniques as well [50]. The optical characteristics such as the one presented in Fig. 6
have been found to be ideal for application of these films in solar control glazing [51].

3.5. Effect of air annealing on the photo response curves

Air annealing leads to significant changes in the chemical, structural, electrical, and
optical properties of chemically deposited semiconductor thin films. Notable exam-
ples are the following:
CdSe thin films: The as-prepared thin films possess very poor photosensitivity:
S"(I !I )/I , where I and I are the photocurrent and dark current, respectively.
1) $ $ 1) $
The CdSe thin films (&0.18lm thickness) in Fig. 7 were deposited at room temper-
ature using N—N,dimethylselenourea as the source of selenide ions [33,34]. The
photosensitivity increases from about 0.5 in as-deposited films to approximately 106
after the film has been annealed in air at 375—425°C for 1 h. This increase in
$&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
Fig. 6. Transmittance (¹%) and near-normal reflectance (R%) curves of chemically deposited CuS thin
films and Cu Se thin films, indicating the high near-infrared reflectance and low infrared transmittance
x
caused by the high electrical conductivities (&103 )~1 cm~1) of the films shown in Fig. 5. Arrows indicate
the positions of absorption edges reported in Ref. [43] for the materials of compositions shown.
324 P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344

photosensitivity arises from the combined effect of grain size growth and chemisorp-
tion of oxygen in the intergrain region as discussed in earlier papers [33,34]. The
mechanism of enhancement of charge carrier mobility in polycrystalline thin-film
semiconductors as a function of grain size and illumination have been the subject of
many earlier studies [52,53]. In short, the chemisorbed oxygen at the intergrain region
builds up repulsive potential for the transport of electrons, leading to very low
electron drift mobilities and hence produces only very small electric currents. Illu-
mination produces electron—hole pairs, of which the holes neutralize the repulsive
potential at the grain boundaries and leads to an exponential increase in the electron
mobility. This results in a high photocurrent under illumination which combined with
the very low current in dark produces a high photosensitivity in the films.
CdS thin films: In sharp contrast with the effect of air annealing on the photo-
sensitivity of CdSe thin films, the photosensitivity of the CdS thin films degrades
under the same annealing process. Fig. 8 shows that the photosensitivity of as-
prepared films is nearly 108, but annealing at temperatures 300—400°C lowers it to
103—104. Annealing at higher temperatures improves the electrical conductivity drasti-
cally, as reported previously [31]. The increase in conductivity arises from the
formation of a top layer of CdO, which is highly n-type due to incomplete oxidation

Fig. 8. Effect of annealing in air for 1 h each at different temperatures of a CdS thin film (200 nm) on the
photocurrent response curves; the high electrical conductivities (&50 )~1 cm~1) of the film obtained after
annealing at 500°C results from a partial conversion of the CdS films to a nonstoichiometric CdO film.
1~x
P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344 325

which renders the CdO layer excess in Cd. X-ray photoelectron spectroscopy studies
indicated that the CdO surface layer resulted from the CdS film [31], and the X-ray
diffraction studies showed that the conversion could not be completed during the air
annealing process — the presence of the underlying CdS film could be detected along
with the CdO component. Assuming that the CdO film thickness is of about 100 nm
(over the CdS film) in the case of the film annealed at 500°C in Fig. 8, the electrical
conductivity is seen to be approximately 100 )~1 cm~1. Same range of conductivity is
observed in CdO films produced by the oxidation of chemically deposited Cd(OH)
2
films [54]. Annealing the CdS films, prepared by the citrate method, in an inert
atmosphere is yet to be studied. It may be possible to identify the optimum annealing
conditions which would result in improvement in the grain size and hence enhance the
photoconductivity in these films without decomposition to oxide.
ZnS thin films: Fig. 9 shows that the ZnS thin films of about 200 nm deposited at
25°C are very resistive; sheet resistance is of nearly 1012 )/h and electrical conductiv-
ity (dark/photo) is nearly 10~8 )~1cm.~1 Air annealing for 1 h at 388°C increases the
photosensitivity to K 104 and annealing at 450°C for 1 h increases the photoconduc-
tivity to '0.05 )~1 cm~1. The formation of a ZnO layer has been indicated by X-ray
diffraction studies. The formation, however, is not complete. XRF studies have shown

Fig. 9. The conversion of as-deposited, nonphotosensitive, ZnS thin film (200 nm thickness) to photo-
sensitive thin film by air annealing at 375—388°C, beyond which the films are partially converted to
nonstoichiometric ZnO thin film, leading to electrical conductivities '0.01 )~1 cm~1.
326 P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344

the presence of about 30% of the initial sulfur content in the films after they were
annealed at 450°C—500°C for up to 2 h. Thus, the ZnO is formed as a surface layer.
The high conductivity arises from the incomplete oxidation, ZnO , similar to the
1~x
case of CdO films. Fig. 10 shows that the ZnS—ZnO thin film maintains a high optical
transmittance &80% in the near-infrared region and shows a shift in the optical band
gap toward lower values. The a2 versus hl plots for the film indicate band gap values
of about 3.7 eV for the as-prepared and 3.4 eV for the ZnS—ZnO film produced by air
annealing at 500°C for 1 h.

Fig. 10. The optical transmittance and reflectance spectra of a chemically deposited ZnS thin film, showing
that the as-prepared and annealed (ZnS/ZnO) films possess high optical transmittance (¹ .%);
#033
¹ .(%)"100 T%/(100 R%).
#033
P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344 327

Fig. 11. Conversion of chemically deposited SnS thin films (250 nm) to SnO films (400—500°C) by air
2
annealing.

SnS thin films: The as-prepared films in this case are very resistive. Annealing at
about 300°C leads to the partial conversion of SnS to SnO ; the nonstoichiometry
2~x
leads to high electrical conductivities. The proportion of SnO in the film increases
2~x
with temperature. Fig. 11 shows that at 400°C the conductivity of the film is the
highest, about 0.5 )~1 cm~1. At this annealing temperature the SnS thin film, which
originally appears orange-red in transmitted daylight becomes nearly transparent
[55]. Annealing at temperatures above 400°C retains the film transparent, but leads to
a better stoichiometry, SnO , which increases the resistance of the films.
2

3.6. Conversion of chemically deposited films to n-type by ion exchange reactions and
diffusion of indium

In earlier papers we have discussed that immersion of CdS thin films in a dilute
solution (0.05 M) of HgCl for a few minutes and annealing in air at 200°C for a few
2
minutes produces n-type conductivity of &1 )~1 cm~1 in the films [31]. This
conductivity is stable during storage in desiccator over a period of several months.
Similar improvements in electrical conductivity are observed in CdSe thin films as
328 P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344

Fig. 12. Conversion of chemically deposited CdSe thin films to n-type through deposition of a 20 nm
indium film on the CdSe film (150 nm) and subjecting it to air annealing at 325°C for 1 h. A chemical
etch for 6 h in a 1 M HCl solution removes the top In O layer, which exposes the underlying
2 3
CdS : In layer.

well [56]. The increase in conductivity has been attributed to the incorporation of
chlorine in the film. The drawback of this type of conversion to n-type is loss of the
conductivity at higher annealing temperatures (300°C or above).
Recently, we reported another technique to convert the chemically deposited
CdS [57,58] and CdSe thin films to n-type [59]. Fig. 12 illustrates the basic
mechanism involved. A 20 nm thick indium film is evaporated over a chemically
deposited CdSe thin film of about 150 nm thickness [59]. The quantity of indium
required for this is about 15 mg to cover an area of 150 cm2 of the film
surface. Annealing at 300—350°C produces an In O surface layer, which may
2 3
be removed by etching in 1 M HCl solution. The underlying film is indium-
doped CdSe, i., CdSe : In. Fig. 13 shows that the conductivity of such a film is high:
nearly 1)~1 cm~1 as compared to that of the CdSe film (10~7 )~1 cm~1) annealed in
air under the same conditions. The conductivity was found to be stable during
desiccated storage.
Fig. 14 shows that thermal diffusion of indium does not inevitably produce n-type
conductivity by substitution of a divalent metal in chemically deposited thin films. In
the case of a 20 nm indium film deposited over a PbS thin film, metallic lead is
P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344 329

Fig. 13. Photocurrent response of annealed CdSe—In films, showing that a conductive In O layer is
2 3
formed on the film, which can be etched away in 1 M HCl solution. The underlying film after a 6 h etch is
devoid of the In O film (Fig. 12), but possesses a dark conductivity of 3 )~1 cm~1, nearly seven orders of
2 3
magnitude higher than that of a CdSe film annealed under the same conditions.

precipitated in the medium, resulting from a reaction of the type


yPbS (crystalline)#x InPIn S (amorphous)# yPb (crystalline).
x y
This reaction takes place when the films are annealed at 300°C, and produces
a composite material. The conductivity of this film is high: 500 )~1 cm~1 and sheet
resistance is about 100 )~2.

3.7. Conversion of chemically deposited thin films to p-type by ion exchange reactions
and interfacial diffusion

The conversion of chemically deposited CdS thin films to p-type by topotaxial


reaction in a CuCl solution of pH 3.7 at 90°C for about 10 s has been developed as
a method to produce CdS—Cu S solar cells in the 1980s [60]. The incorporation of
2
copper ions in chemically deposited CdSe thin films by immersion in a dilute solution
of CuCl (0.005 M) has been reported before [56] from our group. The presence of
2
Cd Cu Se composition in the surface layer, as evidenced in the XPS studies, gives rise
x y z
to very low sheet resistance (and of p-type conductivity) of 105 )/h as compared to
1012 )/h of the CdSe films. The same is the case with ZnSe thin films as well:
330 P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344

Fig. 14. XRD patterns illustrating that annealing in nitrogen of a PbS (200 nm) — In (20 nm) film results in
the formation of metallic lead crystallites and the dispersion of indium in the matrix; the presence of which is
detected in the X-ray flourescence spectra.

immersion for about 1 min of a ZnSe thin film in a 0.025 M CuCl solution reduces
2
the sheet resistance from about 1012 )/h to 103 )/h through the formation of
a Zn Cu Se composition on the surface layer [61].
x y z
Production of thermally stable thin films of p-type conductivity of about
300 )~1 cm~1 (sheet resistance of about 100 )/h) is possible through interfacial
diffusion of metal atoms in ZnS—CuS or PbS—CuS multilayers [5]. The films are stable
in the temperature range 150—350°C under air annealing. More interesting results
were obtained when Bi S —CuS films were annealed. The formation of a new com-
2 3
pound Cu BiS (wittichenite) was suggested in a previous paper [62] and detailed
3 3
results are presented in a recent paper [6]. Fig. 15 illustrates the mechanism of
formation of the new film. Annealing the Bi S —CuS films at temperatures of
2 3
200—300°C results in the interfacial diffusion of metal ions and formation of the new
compound through the reaction

Bi S #6CuSP2Cu BiS #3S.


2 3 3 3
The excess sulfur may be removed by immersion in a suitable solvent. The conductiv-
ity of the resulting film is '100 )~1 cm~1. This indicates that many chemically
deposited thin films involving Cu Se, CuSe, CuS and Cu S may be subjected to
2~x x
suitable annealing process to produce multinary compounds.
P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344 331

Fig. 15. The formation of a new material, Cu BiS (mineral sample; wittichenite) during annealing at
3 3
200—280°C of a chemically deposited Bi S —CuS thin film (X-ray source, Cu K ).
2 3 a

3.8. Use of the precipitate as a precursor for other coating techniques

The real challenge in the optimization of a chemical bath deposition technique is to


reduce the particulate precipitate in the bath obtain in high thin-film yield. Neverthe-
less, even in the best optimized bath, precipitate presents itself as a major product of
the condensation process. The precipitate may be filtered and reacted with acids to
produce the starting materials for deposition. But, in many cases the precipitate may
be rinsed well, dried and stored to serve as precursor for other deposition techniques.
Following are the three specific examples:
(i) Screen printing technique: The basic principles involved in screen printing pro-
cess for producing semiconductor coatings have been discussed in Ref. [63]. This
technique has been successfully used in the production of CdS—CdTe solar cells [64].
The screen printing process involves the preparation of a paste containing the
semiconductor pigment, a flux material which will fuse at a temperature much below
the melting point of the pigment, and a binder usually ethylene glycol or propylene
glycol. The paste is printed on suitable substrates using a silk screen or polyester
screen, dried for a few hours at 50—80°C, and then sintered at a temperature higher
than the melting point of the flux. Fig. 16 illustrates the physicochemical changes
that occur during the sintering process [65]. Here, CdSe precipitate recovered
from a chemical deposition bath [33,34] was mixed with 30% by weight of ZnCl
2
332 P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344

Fig. 16. X-ray (Cu K ) diffraction pattern of a screen-printed and sintered (450°C) CdSe/ZnO layer
a
prepared using CdSe precipitate and 30% by weight of ZnCl as flux (flux-to-pigment ratio, FPR"0.3)
2
dispersed in propylene glycol.

(anhydrous) and sufficient quantity of propylene glycol to produce a screen printing


paste. The paste was screen printed on to glass substrates using a 120 T polyester
screen (approx. mesh: sieve opening 44 lm and filament diameter 30 lm). The coating
was dried in an air oven at 110°C for 1 h and sintered at 350—450°C in air for 1 h each.
The XRD pattern in Fig. 16 for the coating sintered at 450°C shows well-defined
peaks of CdSe and ZnO. A possible mechanism is that at temperatures above the
melting point of ZnCl , it melted and covered the CdSe pigment. At the sintering
2
temperature ZnCl combined with atmospheric oxygen forming ZnO. XRF studies
2
have indicated that the chlorine content of the film has been significantly reduced in
the sintered film — up to 10% of its original value in the dried screen printed coating.
Fig. 17 illustrates the high photosensitivity achieved in the screen printed film
prepared as above. The best photosensitivity, &108, is obtained for a sintering
temperature of 450°C. Photosensitivity of even higher than this (for example, 109
under air mass one solar radiation) has been obtained in chemically deposited CdS
thin films [66,67]. However, the fast decay of photocurrent of the screen printed
CdSe/ZnO coatings is not matched by them.
(ii) ºse in the production of composite coatings: Fig. 18 shows the sheet resistance of
composite coatings made of CuS pigments precipitated from a chemical bath used for
thin-film deposition [42] dispersed in poly(acrylic acid) (molecular weight assay
P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344 333

Fig. 17. Photocurrent response (10 s dark, 10 s illumination, and 10 s dark) of the screen-printed
CdSe/ZnO coating as a function of sintering temperature.

Fig. 18. Sheet resistance of CuS—polyacrylic acid composite coatings, where the CuS precipitate was
obtained from a chemical deposition bath. The CuS weight percentage indicates the percentage weight of
CuS component in the total weight of the composite, poly (acrylic acid) plus the precipitate.
334 P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344

Fig. 19. Variation of sheet resistance in CuS—poly acrylic acid composite coatings with temperature of
annealing.

90 000), with propylene glycol used as a solvent. The conductivity percolation is seen
at 40% w/w of CuS powder in the total weight of the composite. Sheet resistance of
nearly 100 )/h is obtained at a wt% of about 55% CuS. Higher wt% of the
precipitate makes the surface of the coating rough. The results shown are for coatings
annealed at 200°C. Fig. 19 shows that the conductivity does not degrade considerably
when annealed at a temperature up to 250°C. Above this, there is chemical conversion
of the CuS component to digenite, Cu S , (JCPDS 26-0476) as illustrated in Fig. 20
9 5
and a transition of the coating to an insulator. The conversion may take place through
the release of sulfur into the polymer matrix, for example, 15CuSP6CuS#Cu S #4S.
9 5
The resulting changes in chemical bonding have been studied by FTIR and discussed
in Ref. [68].
(iii) ºse of the precipitate as the source for vapor phase deposition: The semiconduc-
tor precipitate produced in the chemical deposition bath is, in general, stoichiometric.
The purity of the precipitate is superior to that of the starting chemicals in the bath
because the impurities whose concentrations are below those required for precipita-
tion into solid phase (i., ionic product(solubility product) are left behind in the
solution. Thus, the precipitate can serve as a relatively pure source of semiconductor
material for vapor phase deposition. In a recent study we used the bismuth sulfide
precipitate recovered from a bath [39] normally used for depositing thin films. The
dried powder was placed in the center zone of a tubular vacuum furnace (10 mTorr)
and heated to 500°C. Glass substrates were placed along the axis of the tubular
furnace from the center of the furnace downstream toward the pumping outlet, with
P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344 335

Fig. 20. Partial conversion of the CuS (covellite) precipitate in the composite coatings to Cu S (digenite)
9 5
when the coatings are annealed at 300°C.

the temperature dropping to nearly 150°C at the far end. The geometry of the vapor
phase deposition set up is discussed in Ref. [69].
Fig. 21 shows that the film deposited at a substrate temperature of 345°C during 5 h
from the bismuth sulfide precipitate maintained at 500°C presents well-defined peaks.
Even though there is a general agreement of the peak positions with that of standard
Bi S (bismuthinite, JCPDS 17320), the intensity ratios observed in the film vary from
2 3
that of the powder sample. This indicates preferential orientation of the crystallites.
Fig. 22 shows the plot of the square of the product of the optical density
(o.d)"(ad)/2.303 (where a is the absorption coefficient and d the thickness of the film)
and the photon energy E"hc/j (where h is the Planck’s constant; c the velocity of
light; and j the wavelength of radiation). The thin films possess a wide range of direct
optical band gaps from 1.0 to 2.0 eV, depending on the temperature of the substrate.
The range of photosensitivity, S"(I !I )/I "(p !p )/p (where I and p are
1) $ $ 1) $ $ 1) 1)
the photocurrent and photoconductivity, respectively, and I and p are the dark
$ $
current and dark conductivity, respectively) varies widely as shown in Fig. 23. High
electrical conductivity is typical of films deposited at higher substrate temperatures.
Thus, the optical and electrical properties of the films may be varied depending on the
chosen substrate temperature to suit specific applications of the films.

3.9. Chemically deposited semiconductor thin films for solar energy related applications

Major advantages of the chemical bath deposition for solar energy applications: Two
major characteristics of solar energy are well recognized: (i) It is abundant in most part
336 P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344

Fig. 21. XRD patterns of bismuth sulfide thin films deposited with substrate temperature of 345°C using
bismuth sulfide precipitate as source, source temperature, 500°C at 4000 Pa. Standard patterns for Bi S
2 3
(bismuthinite) is given for the sake of comparison.
P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344 337

Fig. 23. Photocurrent response of bismuth sulfide thin films deposited from vapor phase at different
substrate temperatures using bismuth sulfide precipitate as the source. The values of photosensitivity
S are given.

of the world; and (ii) it is available in a relatively dilute form, typically 20 MJm~2/day,
as compared with nearly 30 MJ/kg obtainable from coal or petroleum. Collection
of solar energy over a large area is inevitable. Chemical bath deposition is considered
to be excellent in producing semiconductor thin films over large areas for such
applications.
(i) Application in solar control coatings: In previous papers we have described the
application of various chemically deposited semiconductor thin films for this purpose
[12,13]. The basic purpose of this application is to reduce space cooling expenses in
buildings and in automobiles by selectively controlling the amount of visible and
infrared radiation entering through glazings. Chemically deposited CuS thin films
have been found to possess near-ideal solar control characteristics: transmittance in
the visible region of 20—50%, low transmittance, 10—20%, in the infrared region, low
reflectance, (10%, in the visible region so as to avoid glare, and relatively higher
reflectance, '15%, in the near-infrared region [70]. The coatings can be made on
$&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
Fig. 22. Variation in the optical band gap of bismuth sulfide thin films deposited at different substrate
temperatures (205—380°C) from vapor phase using bismuth sulfide precipitate as the source (source
temperature, 500°C; pressure: 4000 Pa).
338 P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344

sheet glass, acrylic, or polyester sheet and foils with equal ease [71]. To impart
versatility and improvement of adhesion of the film to glass substrate, substrate films
of chemically deposited PbS [72], SnS [73], ZnS [35], Bi S [74], and ZnS—NiS [75]
2 3
have been proposed. A major problem in the application of these films was the need
for a protective coating. The use of an aromatic polyurethene-based commercial
transparent varnish was tried with limited success [76].
We have recently concluded that the best way to use chemically deposited
solar control coatings is in the form of a laminated glazing. In this configuration,
coatings were made on 3 mm thick commercial sheet glass and then laminated
using a commercially available laminating polymer sheet of 0.4 mm thickness.
This is a proprietary copolymer [77] of polyvinyl butyryl-acetate — alcohol with
other ingredients added in order to prevent degradation of the polymer by
the ultraviolet component of solar radiation. The lamination, of glass # coating—
polymer—clear sheet glass, was done in an autoclave at 140°C under 2 kg cm~2
pressure for about 3 h.
Fig. 24 illustrates the transmittance and reflectance spectra of two different coatings
CuS and PbS-CuS, where the individual coating thicknesses are of the order of 0.1 lm.
The spectra were recorded for incidence from the side of the coated glass — the
recommended mode of incidence to minimize entry of the ultraviolet component of
solar radiation into the polymer. The calculation of integrated reflectance (R) and
transmittance (¹) in the visible (vis) and infrared (ir) were calculated for air mass two
solar spectrum as described in Ref. [12]. The implication of these values in the solar
control parameters of the glazing is discussed in Ref. [13]. The visible transmittance,
¹ , of 51% and visible reflectance, R , of 12% of the laminated CuS coatings are
7*4 7*4
suitable for use in automobile glazing and combination of ¹ "21%, R "18%,
7*4 7*4
and R "28% of the laminated PbS—CuS coating are suitable for architectural
*3
glazing applications.
(ii) Application as solar absorber coatings: Chemically deposited SnS—CuS [78],
PbS—CuS and Bi S —CuS [79] coatings have been investigated by us for application
2 3
as absorber coating in all-glass tubular solar collectors. The efficiency of such
coatings has been analyzed in a theoretical model [11]. It was apparent that
the absorber coating must be located on the outside of the inner glass tube. The
stability of the coating in the vacuum zone at operating temperature of 300°C
is therefore the major issue to be addressed. In the case of PbS—CuS and ZnS—CuS
thin films, thermal stability at temperatures of up to 350°C has been observed
to be satisfactory during air annealing [5]. The solar absorptance of a PbS—CuS
film is about 0.87 and thermal emittance is less than 0.46 (corresponding to sheet
resistance of the annealed film of about 32 )h [79]. As per theoretical prediction
[11], such a coating can offer conversion efficiency of about 50% at an f-value
of 0.1°C m2 W~1. This value is not very attractive compared to what may be
offered by coatings produced by vacuum techniques [80]. However, the advantage
of chemical bath deposition in producing uniform coatings on glass tubes is substan-
tial over vacuum techniques. Further work on the thermal stability of chemically
deposited multilayer films in vacuum at 300—400°C is worth considering to develop
this application.
P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344 339

Fig. 24. Optical transmittance and reflectance of CuS and PbS—CuS coatings deposited on 3 nm thick
commercial window glass and laminated with a clear window glass (3 mm) using commercial-grade
laminating polymer [77]. Laminating process: 140°C under 2 kg/cm2 pressure for 2 h.

(iii) Photodetector and photovoltaic applications: Fig. 25 shows the dependence of


the photoresponse curve of chemically deposited CdSe thin films as a function of the
intensity of illumination. The dependence of the photocurrent on the intensity is
nearly linear. The photo generation of such magnitude in the depletion layer of
a Schottky barrier or p—n, p—i—n junction leads to the build up of photo voltage. We
consider that the possibility of creating new materials through interfacial diffusion of
atoms in chemically deposited multilayer stacks as demonstrated in PbS—CuS,
ZnS—CuS and Bi S —CuS films [5,6] is very high. Table 1 suggests a list of possible
2 3
absorber materials for photovoltaic application chosen from the data given in
340 P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344

Fig. 25. Photoresponse curves of a CdSe thin film annealed in air at 450°C at different illumination levels
(0.01%), 0.2 W m~2 and to 100%, 2.12 kW m~2 illustrating near-linear dependence of photocarrier
generation on the intensity of illumination — a basic feature on which photovoltaic effect depends.

Table 1
Suggestions for the development of new solar cell/solar energy materials which may be prepared through
the interfacial diffusion of ions in multilayer chemically deposited thin films during annealing. The data
have been extracted from Ref. [43]

Material Structure Band gap (eV) Hole mobility


(cm2/Vs)

CuSbSe Orthorhombic 0.83 5


2
AgSbS Monocl. 1.73 1500
2
AgSbSe Rock salt 0.58
2
AgBiS Rock salt 0.9
2
Cu SbS Zincblende 0.46
3 4
Cu SbSe Zincblende 0.31 50
3 4
Ag SnSe 0.81 910
2 3
Cu SnSe 0.96 870
2 3
Cu SnS 0.91 605
2 3
PbSnS 1.05
3
P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344 341

Ref. [43], which may be produced by solid-state reaction of the type


2CuS#SnSPCu SnS (E "0.91 eV),
2 3 '
PbS#SnS PPbSnS (E "1.05 eV),
2 3 '
Bi S #Ag SP2AgBiS (E "0.9 eV),
2 3 2 2 '
Sb S #Ag SP2AgSbS (E "1.73 eV),
2 3 2 2 '
etc., where the CuS [42], SnS [37], PbS [38], Bi S [39] Ag S [81] Sb S [48] films
2 3 2 2 3
are produced by chemical bath deposition. In the case of SnS and SnS , it has been
2
reported that by controlling the bath composition and temperature either of the
composition may be obtained [82]. Further, new materials may also be created by
combining chemical deposition technique with thermal evaporation or sputtering of
metals, for example, Sb S #3CuS#3CuP2Cu SbS , etc. We consider that cre-
2 3 3 3
ation of new thin-film materials by the above approaches is vital for the further
advancement of the chemical bath deposition technique and integration of the films
produced in this way into solar energy technologies.

3.10. Toxicity considerations

The chemical deposition technique involves the use of dilute solutions of com-
pounds involved in the reaction. This offers minimum toxicity and occupational
hazards since the vapor phase of the reactants are avoided. It is well known that
toxicity hazards associated with lead, cadmium, mercury, selenium, etc, are severe
when inhaled [83]. An analysis on the toxicity hazard in the production of chemically
deposited PbS thin films is given in Ref. [84] where it was established that the
chemical bath technique does not entail the usual health hazards associated with lead
pollution. Further, the unreacted ions can be precipitated in the bath as sulfides or
selenides and the solid can be separated and stored for use as discussed above in
Section 3.8 or recycled to produce starting material. The formation of multinary
compounds by interfacial diffusion and recrystallization in multilayer films as given
above can be considered as an environmentally sound process since very few effluents
are produced. Overall, the-large area capability and the ease of scaling up with
complete control of material handling in solid or liquid phase offers perspective
toward the industrial production of coatings and devices by the chemical bath
deposition technique.

4. Conclusions

In this paper we presented a perspective on the chemically deposited thin films: the
deposition technique, post-deposition processing, the use of precipitate and an over-
view of the applications related to solar energy utilization. We consider that this
research area offers immense possibilities in creating new materials by annealing
multilayer film or combining chemically deposited films with vacuum-deposited
metallic films. Future research efforts may be directed in this direction.
342 P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344

Acknowledgements

The authors are grateful to Leticia Ban8 os of the Instituto de Investigaciones en


Materials, UNAM, for recording the XRD patterns. Financial support for this work
was received from CONACYT, México through 1816-E9211 and from DGAPA-
UNAM through IN-502594 and IN-501495. One of us (V.M.G) is grateful to the
Universidad Autonoma de Zacatecas for financial support.

References
[1] O. Houser, E. Beisalski, Chem-Ztg. 34 (1910) 1079.
[2] K.L. Chopra, R.C. Kainthla, D.K. Pandya, A.P. Thakoor, in: G. Hass, M.H. Francombe, J.L. Vossen
(Eds.), Physics of Thin Films, vol. 12, Academic Press, New York, 1982, p. 201.
[3] C.D. Lokhande, Mater. Chem. Phys. 27 (1991) 1—43.
[4] I. Grozdanov, Semicond. Sci. Technol. 9 (1994) 1234.
[5] L. Huang, P.K. Nair, M.T.S. Nair, R.A. Zingaro, E.A. Meyers, J. Electrochem. Soc. 141 (1994) 2536.
[6] P.K. Nair, L. Huang, M.T.S. Nair, Hailin Hu, E.A. Meyers, R.A. Zingaro, J. Mater. Res. 12, 1997,
(in Press).
[7] D.E. Bode, in: G. Hass, R.E. Thun (Eds.), Physics of Thin Films, vol. 3, Academic Press, New York,
1966, p. 275.
[8] A.G. Stanley, Cadmium Sulfide Solar Cells, in: R. Wolf (Ed.), Applied Solid State Sciences, vol. 5,
Academic Press, New York, 1975, p. 251.
[9] S.V. Svechnikov, E.B. Kaganovich, Thin Solid Films 66 (1980) 41.
[10] G.B. Reddy, V. Dutta, D.K. Pandya, K. L Chopra, Sol. Energy Mater. 15 (1987) 383; 15 (1987)
153.
[11] C. Estrada-Gasca, G. Alvarez-Garcı́a, R.E. Cabanillas, P.K. Nair, Renewable Energy 2 (1992) 477;
J. Phys. D 25 (1992) 1142.
[12] P.K. Nair, M.T.S. Nair, A. Fernández, M. Ocampo, J. Phys. D 22 (1989) 829.
[13] C. Estrada-Gasca, G. Alvarez-Garcı́a, P.K. Nair, J. Phys. D 26 (1993) 1304.
[14] R.A. Boudreau, R.D. Rauh, J. Elctrochem. Soc. 130 (1983) 513.
[15] M.E. Rincón, M. Sánchez, A. Olea, I. Ayala, P.K. Nair, Photoelectrochemical behavior of thin CdS,
coupled CdS/CdSe semiconductor thin films, Solar Energy Mater. Solar Cells (1997), These pro-
ceedings.
[16] O. Savadogo, Chemically and electrochemically deposited thin films for solar energy materials, Solar
Energy Mater. Solar Cells (1997) These proceedings.
[17] Bluent Basol, Vijay Kapur, IEEE Trans. Electron Dev. 37 (1990) 418.
[18] T.R. Tuttle, M.A. Contreras, J.S. Ward, A.L. Tennant, K.R. Ramanathan, J. Keane, R. Noufi, in: C.M.
Lampert et al. (Ed.), Proc. SPIE, vol. 2531, SPIE, Bellingham, 1995, p. 194.
[19] M.A. Mickelsen, W.S. Chen, Appl. Phys. Lett. 36 (1980) 371.
[20] A. Rothwarf, Proc. 16th IEEE Photovoltaic Specialists Conf., San Diego, CA, IEEE, New York, 1982,
p. 791.
[21] B.M. Basol, V.K. Kapur, A. Halani, C. Leidholm, in: K.A. Summers (Ed.), Annual Report, Photovol-
taic Subcontract Program FY 1991, Golden, NREL, p. 50.
[22] O. Savadogo, K.C. Mandal, Appl. Phys. Lett. 63 (1993) 12.
[23] O. Savadogo, K.C. Mandal, J. Electrochem. Soc. 141 (1994) 2871.
[24] O. Savodogo, K.C. Mandal, J. Phys. D 27 (1994) 1070.
[25] P.K. Nair, V.M. Garcı́a, A.B. Hernandez, M.T.S. Nair, J. Phys. D 24 (1991) 1466.
[26] P.K. Nair, O. Gomezdaza, M.T.S. Nair, Adv. Mater. Opt. Electron. 1 (1992) 139.
[27] P.K. Nair, M.T.S. Nair, O. Gomezdaza, R.A. Zingaro, J. Electrochem. Soc. 140 (1993) 1085.
[28] J. Basset, R.C. Denney, G.H. Jeffery, J. Mendham, in: Vogel’s Textbook of Quantitaive Inorganic
Analysis, 4th ed., ELBS/Longman, London, 1978, p. 25.
P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344 343

[29] D.R. Lide, (Editor-in-Chief), CRC Handbook of Chemistry and Physics 76th Ed., CRC Press, Boca
Raton, 1995-96, pp. 8-58.
[30] J. Bjerrum, B.V. Agarwala, S. Refn., Acta Chem. Scand. A 35 (1981) 685.
[31] N.R. Pavaskar, C.A. Menezes, A.B.P. Sinha, J. Electrochem. Soc. 124 (1977) 743.
[32] M.T.S. Nair, P.K. Nair, R.A. Zingaro, E.A. Meyers, J. Appl. Phys. 75 (1994) 1557.
[33] M.T.S. Nair, P.K. Nair, H.M.K.K. Pathirana, R.A. Zingaro, E.A. Meyers, J. Electrochem. Soc. 140
(1993) 2987.
[34] V.M. Garcı́a, M.T.S. Nair, P.K. Nair, R.A. Zingaro, Semicond. Sci. Technol. 11 (1996) 427.
[35] P.K. Nair, M.T.S. Nair, Semicond. Sci. Technol. 7 (1992) 239.
[36] C.A. Estrada, P.K. Nair, M.T.S. Nair, R.A. Zingaro, E.A. Meyers, J. Electrochem. Soc. 141 (1994) 802.
[37] M.T.S. Nair, P.K. Nair, Semicond. Sci. Technol. 6 (1991) 132.
[38] P.K. Nair, M. Ocampo, A. Fernandez, M.T.S. Nair, Sol. Energy Mater. 20 (1990) 235; V.M. Garcı́a,
M.T.S. Nair, P.K. Nair, Sol. Energy Mater. 23 (1991) 47.
[39] M.T.S. Nair, P.K. Nair, Semicond. Sci. Technol. 5 (1990) 1225.
[40] P.K. Nair, J. Campos, A. Sánchez, L. Ban8 os, M.T.S. Nair, Semicond. Sci. Technol. 6 (1991) 393.
[41] V.M. Garcı́a, M.T.S. Nair, P.K. Nair, R.A. Zingaro, Semicond. Sci. Technol. 12 (1997) (May issue), in
press.
[42] P.K. Nair, V.M. Garcı́a, A.M. Fernández, H.S. Ruiz, M.T.S. Nair, J. Phys. D 24 (1991) 441.
[43] O. Madelung, in: Semiconductors other than Group IV elements, III—V compounds, Springer, Berlin,
1992.
[44] M. Moskovits, in: G. Soles (Ed.), Chemical Physics of Atomic and Molecular Clusters, North Holland,
Amsterdam, 1990, p. 397.
[45] For example, M.N. Rudden, J. Wilson, Elements of Solid State Physics, Wiley, Chichester, 1980, p. 59.
[46] L. Brus, J. Chem. Phys. 90 (1986) 2555.
[47] G. Hodes, A.A. Yaron, F. Decker, P. Motisuke, Phys. Rev. B 36 (1987) 4215.
[48] K.C. Mandal, O. Savadogo, Sol. Energy Mater. Sol. Cells 26 (1992) 117.
[49] K. Okamoto, S. Kawai, Jpn. J. Appl. Phys. 12 (1973) 1130; J.J. Loferski, J. Schewchun, S.D. Mittleman,
E.A. DeMeo, R. Arnott, H.L. Hwang, R. Beauliew, G. Chapman, Sol. Energy Mater. 1 (1979) 157.
[50] H.S. Randhawa, R.F. Bunshah, D.G. Brock, B.M. Basol, O.M. Staffsudd, Sol. Energy Mater. 6 (1982)
445.
[51] M.T.S. Nair, P.K. Nair, Semicond. Sci. Technol. 4 (1989) 191.
[52] J.W. Orton, B.J. Goldsmith, J.A. Chapman, M.J. Powell, J. Appl. Phys. 53 (1982) 1602.
[53] M.V. Garcı́a-Cuenca, J.L. Morenza, J.M. Cordina, J. Phys. D 20 (1987) 951.
[54] R.L. Call, N.K. Jaber, K. Seshan, J.R. Whyte, Sol. Energy Mater. 2 (1980) 373.
[55] P.K. Nair, M.T.S. Nair, R.A. Zingaro, E.A. Meyers, Thin Solid Films 239 (1994) 85.
[56] M.T.S. Nair, P.K. Nair, H.M.K.K. Pathirana, R.A. Zingaro, E.A. Meyers, J. Electrochem. Soc. 140
(1993) 2987.
[57] P.J. George, A. Sánchez, P.K. Nair, M.T.S. Nair, Appl. Phys. Lett. 66 (1995) 3624.
[58] P.J. George, A. Sánchez-Juarez, P.K. Nair, Semicond. Sci. Technol. 11 (1996) 1090.
[59] V.M. Garcı́a, P.J. George, M.T.S. Nair, P.K. Nair, J. Electrochem. Soc. 143 (1996) 2892.
[60] M. Savelli, J. Bougnot, in: B.O. Seraphin (Ed.), Solar Energy Conversion and Topics in Applied
Physics, vol. 31, Springer, Berlin, 1979, p. 213.
[61] C.A. Estrada, R.A. Zingaro, E.A. Meyers, P.K. Nair, M.T.S. Nair, Thin Solid Films 247 (1994) 208.
[62] P.K. Nair, M.T.S. Nair, H.M.K.K. Pathirana, R.A. Zingaro, E.A. Meyers, J. Electrochem. Soc. 140
(1993) 754.
[63] K.L. Chopra, S.R. Das, in: Thin Film Solar Cells, Plenum, New York, 1982, p. 233.
[64] H. Matsumoto, K. Kuribayashi, H. Uda, Y. Komatsu, A. Nakamoto, S. Ikegami, Sol. Cells 11 (1984)
367.
[65] O. Gomezdaza, V.M. Garcı́a, M.T.S. Nair, P.K. Nair, Appl. Phys. Lett. 68 (1996) 1987—1989.
[66] P.K. Nair, M.T.S. Nair, J. Campos, L.E. Sansores, Sol. Cells 22 (1987) 213.
[67] M.T.S. Nair, P.K. Nair, J. Campos, Thin Solid Films 161 (1988) 21.
[68] H. Hu, J. Campos, P.K. Nair, J. Mater. Res. 11 (1996) 739.
[69] M.E. Rincon, P.K. Nair, Semicond. Sci. Technol. 12 (1997) at Press.
344 P.K. Nair et al. /Solar Energy Materials and Solar Cells 52 (1998) 313—344

[70] M.T.S. Nair, P.K. Nair, Semicond. Sci. Technol. 4 (1989) 599.
[71] H. Hu, P.K. Nair, Surface Coating Technol. 81 (1996) 183.
[72] P.K. Nair, M.T.S. Nair, Semicond. Sci. Technol. 4 (1989) 807.
[73] M.T.S. Nair, P.K. Nair, J. Phys. D 24 (1991) 450.
[74] M.T.S. Nair, G. Alvarez-Garcia, C. Estrada-Gasca, P.K. Nair, J. Electrochem. Soc. 140 (1993) 212.
[75] A.M. Fernandez, M.T.S. Nair, P.K. Nair, Mater. Manufr. Process 8 (1993) 535.
[76] A.M. Fernandez, P.K. Nair, Thin Solid Films 204 (1991) 459.
[77] Vitro Vidrio Plano-Industria Quimica M, S.A. de C.V., Tlaxcala-Puebla, Tlaxcala C.P.90780 AP 942
México.
[78] P.K. Nair, M.T.S. Nair, J. Phys. D 24 (1991) 83.
[79] V.M. Garcia, M.T.S. Nair, P.K. Nair, Solar Energy Mater. 23 (1991) 47.
[80] B. Window, Solar Energy 31 (1983) 159.
[81] A.J. Varkey, Solar Energy Mater. 21 (1991) 291.
[82] R.D. Engelken, H.E. McCloud, C. Lee, M. Slayton, A. Ghreishi, J. Electrochem. Soc. 134 (1987) 2696.
[83] N.I. Sax, Dangerous properties of Industrial Materials, 6th ed., Van Nostrand, Reinhold, New York,
1984, p. 612, 1693, 1751, 2390.
[84] P.K. Nair, M.T.S. Nair, J. Phys. D 23 (1990) 150.

You might also like