CHAPTER - III

3 Chemical bath deposition and characterization of CdS films

3.1 CdS : Introduction

Cadmium sulfide is a chemical compound with the molecular formula CdS.

Cadmium sulfide is yellow in colour and is a semiconductor. It exists in nature as two

different minerals, hexagonal greenockite [1] and cubic hawleyite [2]. Cadmium

sulfide is a direct band gap semiconductor had energy 2.42 eV [3] and has many

applications for example in light detectors. It forms thermally stable pigments and

with the addition of CdTe and HgS, the colours of the pigments changes from deep

red to yellow [4]. Figure 16 shows the molecular structures (greenokite and hawleyite)

of CdS.

3.1 (a) Chemical properties

Cadmium sulfide is soluble in acids and this has been investigated as a method of

extracting the pigment from waste polymers [5].

CdS + HCl → CdCl2 + H2S

When sulfide solutions containing dispersed CdS particles are irradiated with light

hydrogen gas is generated. H2S → H2 + S ∆Hf = + 9.4 kcal / mol

The reaction mechanism proposed involves the electron/hole pairs created when

incident light is absorbed by the cadmium sulfide followed by these reacting with

water and sulfide [3, 6].

Production of an electron hole pair CdS + hν → e− + hole+

Reaction of electron 2e− + 2H2O → H2 + 2OH−

Reaction of hole 2hole+ + S2− → S

41
Figure 16 The molecular structures (a) greenokite and (b) hawleyite of CdS.

Table 1 General physical and chemical properties of CdS

Molecular formula CdS

Molar mass 144.46 g / mol
Appearance Yellow-orange solid.
Density 4.82 g / cm3
Boiling point 980 °C
Solubility in water 0.13 g / L (20° C)
Refractive index 2.51

42
3.1 (b) Physical properties

Cadmium sulfide exist in two crystalline forms; the more stable hexagonal

wurtzite structure was found in the mineral Greenockite and the cubic zinc blende

structure was found in the mineral Hawleyite. In both of these forms the cadmium and

sulfur atoms are four coordinate. There is also a high pressure form with the NaCl

rock salt structure [7]. Cadmium sulfide is a direct bandgap semiconductor with a

bandgap of 2.42 eV at 300 K. The magnitude of its band gap means that it appears

coloured [1-3]. The conductivity increases when irradiated with light which leads to

use CdS as a photoresistor. When combined with a p-type semiconductor it forms the

core component of a photovoltaic (solar) cell and a CdS/Cu2S solar cell was one of the

first efficient cells to be reported [8-9]. When doped with for example Cu+

("activator") and Al3+ ("coactivator") CdS luminesces under electron beam excitation

(cathodoluminescence) and is used as phosphor [10]. Both polymorphs are

piezoelectric and the hexagonal is also pyroelectric [11]. CdS crystal can act as a solid

state laser [12-14]. Table 1 illustrates the general physical and chemical properties of

CdS.

There are many methods to fabricate CdS thin films, such as electrodeposition

[15], chemical bath deposition (CBD) [16–20], molecular beam epitaxy (MBE) [21],

metal organic vapor phase epitaxy (MOVPE), metal organic chemical vapor

deposition (MOCVD) [22-23], close spaced sublimation (CSS) [24], successive ionic

layer adsorption and reaction (SILAR) [25], screen printing [26], and physical vapor

deposition (PVD) [27]. Each method has its advantage and disadvantages in device

applications. Among these methods, CBD is a very attractive method to produce CdS

films for photovoltaic applications for which large-area devices and low-cost

43
processes are required. Chemical bath deposition yields stable, uniform, adherent and

hard films with good reproducibility by a relatively simpler process. Especially, the

CBD technique requires very low temperature (less than 90 °C) by Jae-Hyeong Lee et

al. [28-29].

3.2 Literature on chemical bath deposited CdS

More efforts has been dedicated to synthesize nanomaterial CdS having various

morphologies including nanowire, nanorod, nanotube, hollow sphere, peanut and

nanocable by various physical and chemical solution methods [30-34]. A novel

thioglycolic acid (TGA) assisted hydrothermal method to prepare the CdS nanorods

was reported by Hui Zhang et al [35-37]. This method is milder, simpler, more

practical, and more environmental friendly than other methods. One can modify the

shape of the CdS nanostructure by adjusting the four critical factors including

hydrothermal temperature, capping molecules, temperature ramp-up rate and type of

sulfur source. Here, TGA as the capping molecules are beneficial for the formation of

CdS nanorods having different morphology. In particular, TGA acts as the oriented

growth reactant.

The controllable growth of flower - like CdS nanostructure by organic-free

hydrothermal process using thioacetamide (TAA) as sulfur source was also reported

[38]. The flower-like CdS nanostructures made up of sword - like CdS nanorods are

single crystalline and hexagonal phase. The internal structure and the secondary

growth are mainly controlled by sulfur mass transport during the synthesize of the

flower - like CdS nanostructures. Sodium sulfide and TAA were used as sulfur source.

Compared to sodium sulfide, the action of TAA on the formation of flower - like CdS

nanostructures was more.

44
 Rapid formation of CdS and ZnS thin films by microwave-assisted chemical bath

deposition technique was reported by Ran Zhai et al [39]. Microwave heating is a

quite fast, simple and efficient method to prepare inorganic materials. The exact

nature of microwave interaction with reactants during the synthesis of materials is

somewhat unclear and speculative. However, energy transfer from microwaves to the

material is believed to occur either through resonance or relaxation [40]. The

deposition time of the CdS and ZnS films by the microwave assisted – chemical bath

deposition is only few 10 minutes duty cycle; it is obvious that using the microwave

assisted - chemical bath deposition method to fabricate the metal sulfide films is more

efficient in terms of power consumed, time saved and property of the film owing to

the different heating mechanism involved when compared to conventional bath

heating.

Yonghong Ni et al [41] have designed a novel microwave - assisted template

synthesis route to prepare an ordered array of CdS nanotubes. CdCl2 and thiourea

were used as cadmium ion and sulfur ion sources, respectively. The formation process

of CdS nanotubes contains two stages: firstly, the CdS sol was obtained by microwave

heating and then, the CdS nanoparticles deposited on the pore walls of porous anodic

alumina membrane to form nanotubes. The tubular structure of the product was

revealed by TEM and SEM observations. This new method is very useful to

synthesize the ordered array of semiconductor sulfide nanotubes.

Synthesis and characterization of various sized CdS nanocrystals in pore of

polyacrylamide gel at room temperature was reported [42]. It is well known that

polyacrylamide gel is a three dimensional network structure and with increasing of the

cross-linking density, the pore size of the gel will decrease [43]. The prepared size of

45
the CdS nanocrystals in the pore of the polyacrylamide gel was controlled by the pore

size of the gel. This will offer a facile method to prepare desired sized semiconductor

nanoparticles in gel medium.

Synthesis of cadmium sulfide nanoparticles via thermal destruction of thiourea

complexes was studied [44]. This approach was prompted by the fact that transition -

metal complexes with thiourea as a sulfiding agent decompose on heating to form the

corresponding metal sulfides. Metal salts begin to react with thiourea (sulfiding agent)

in solution, resulting in Cd and S coordination in the complex, which ensures the

introduction of sulfur atoms into the nearest neighbor environment of the metal atoms.

This leads to the formation of structurally perfect sulfides.

A novel CdS micro – spheres, resembling soccer and consisting of CdS

nanocrystals with wurtzite structure were synthesized via a simple chemical bath

deposition method conducted at 90 °C for 24 h [45]. In addition, the formation

mechanism of this special shape was also discussed. Even though, the chemical bath

was kept at different temperatures ranging from 40 to 90 °C for 24 h, the reported

FESEM image of the sample synthesized at 90 °C for 24 h, showed a mass of densely

packed spheres with a diameter distribution from 5 to 30 µm. This was attributed to

the formation mechanism involved as well as the growth speed of CdS spheres. In this

experiment thioacetamide was used as the sulfur source. The decomposition of

thioacetamide speds up with the increasing temperature and it was easy to reach the

critical supersaturation of CdS at a higher temperature. The reported results revealed

that the reaction temperature affects the size and the shape of the CdS spheres.

The results of preparation and characterization of nanocrystalline thin films of CdS

within the self organized pores of polyvinyl alcohol (PVA) via a chemical bath

46
deposition process was reported [46]. It was observed that crystallite size decreases

when the molarity of cadmium sulphate increases in the chemical bath solution.

The effect of polyvinylpyrrolidone (PVP) in the reaction system for the synthesis

of CdS nanowire was investigated and it has been found that 0.8 g of PVP / 50 ml of

water was the best dosage than 0.2 g / 50 ml and 2 g / 50 ml to have CdS nanowires

with uniform and smooth faces [47].

Feitosa et al have reported a new root for the deposition of CdS thin films using

chemical bath deposition technique with EDTA as a ligand [48]. Different

concentrations of EDTA were proved to estimate the structural quality of the CdS thin

films. Furthermore, the variation of pH in the chemical bath during the growth is

important to get good quality films. EDTA is an amino - derived organic -compound

known to be a strong hexdentate chelating agent. It forms a complex with metal ions

and dissociates reversibly at a low rate. Lade and Lokhande [49] have used EDTA in

the electrodeposition of CdS thin films from non-aqueous bath. Furthermore, EDTA

has been used in several process involving CdS colloid [50] or cadmium particles [51-

52]. EDTA affects the adsorption and desorption processes of cadmium in red soil

(ferrisoils), and also, it affects the growth of CdS crystalline thin films by

photochemical deposition [53 - 54].

Jae-Hyeong Lee et al [55] have reported the structural and optical properties of

CdS thin films on glass, polycarbonate (PC), and polyethylene terephthalate (PET).

The CdS polycrystalline films showed hexagonal <0 0 2> / cubic <1 1 1> orientation

when deposited onto glass substrate, but such preferential orientation decreased or

disappeared when the deposition was made onto polycarbonate and

polyethyleneterephthalate substrates. The roughness value of the CdS films was

47
influenced by various substrates. The average transmittance of the films on PC and

PET was about 50% and 55%, respectively, when compared to 70% for glass

substrate. The quality of the CdS films deposited on organic substrates is somewhat

worse than that on glass substrate.

In the present work, attempts are made to construct an experimental set-up for the

chemical bath deposition of CdS films. The experimental parameters such as bath

temperature, substrate etching, duration of the experiment and concentration of CdCl2

and thiourea were considered to investigate their role on the formation mechanism of

CdS thin films and its properties. After the optimization of these experimental

parameters, the CdS samples are subjected for X-ray diffraction, SEM, AFM, UV-

Vis-NIR and room temperature PL emission spectroscopy analysis to investigate the

structural, surface morphology, optical transparency and photoluminescence emission

properties. The results obtained from these studies were summarized and discussed in

the following paragraphs.

3.3 Experimental setup

Figure 17 illustrates the experimental set-up developed in our laboratory to deposit

the chalcogenide family semiconducting materials such as CdS, CdZnS, CuCdS thin

films on various substrates by chemical bath method. Basically the instrument consists

of a glass bath of volume 250 ml and attached with a plate-heater at the bottom. The

bath temperature was maintained at the desired temperature by a constant voltage

regulator. The substrate was suspended vertically inside the chemical bath using

teflon-disc substrate holder and was rotated at 20 rpm. Care was taken to avoid any

mechanical disturbance from the experimental set-up to the growth except substrate

rotation.

48
Figure 17 The photograph of the in-house developed chemical bath

deposition instrument

49
Rotating the substrate facilitates to make the solution more homogenized and to

expose the substrate surface in phase with the growth units available in the growth

solution there by more uniform films can be expected. All the experiments were

conducted over a period of one hour by keeping the chemical bath stationary. The

commercially bought raw materials from M/S.Merck India Ltd., were utilized for this

study. For CdS films, cadmium chloride and thiourea were used as the starting

materials and ammonia as a ligand to control the precipitation of chalcogenides and

hydroxides. Here, a stable CdCl2, which slowly releases the cadmium ions through the

controlled chemical reaction with ammonium hydroxide solution. Thiourea releases

sulphur ions by means of an alkaline hydrolysis process. The chemical reaction

involved during this process is explained as follows;

CdCl2 + 2NH4OH Æ Cd(OH)2 + 2NH4CI

Cd(NH3)]42+ Æ Cd2+ + 4NH3

CS(NH2)2 +2H2O Æ 2NH3 + CO2 + H2S

H2SÆS2- + 2H+

The final product is Cd2+ + S2- ÆCdS.

During the chemical bath deposition of CdS films, it is believed that there are two

possible pathways, i.e., clusters – by – clusters deposition via CdS colloidal particles

formed in the solution, and ion – by – ion deposition via metastable complex

comprising Cd and S agents [56-57]. These two deposition pathways are competitive,

which depends on the effective concentrations of Cd2+, S2-, and CdS particles in the

chemical bath. When the effective concentrations of Cd2+ and S2- are high and that of

CdS particles is low, ions – by – ions deposition is dominant in chemical bath

deposition process; on the contrary, the clusters – by – clusters deposition dictates the

50
chemical bath deposition process. Further, the ions – by – ions deposition gives

crystalline, more adherent and uniform films. In this regard, Mendoza – Perez et al.

have reported that the chemical bath deposited CdS window layer on CdTe layer

exhibit the photovoltaic efficiencies η = 7.1, 8.3 and 12.3 % with respect to S / Cd

ratio of 1 / 1, 2.5 / 1 and 5 / 1, respectively. Also, the best efficiency of 16.5% for a

CdTe solar cell was achieved with a CdS window layer grown by chemical bath

deposition technique by Wu et al. [58-59].

3.4.1 Substrate preparation

The glass substrates of size, 1 x 1 cm2 were etched for 5 minutes in 1% hydro

fluoric (HF) acid solution to improve the adherence of the film. Prior to the

deposition, glass and indium tin oxide (ITO) coated glass substrates were

ultrasonically cleaned for 15 minutes and then kept in acetone – bath. In the case of

silicon <111> and <100> substrates, the as cut and lapped substrates, which are

obtained from commercial vendors were chemically polished using HF, HNO3 and

H2O taken in the volume ratio of 1: 6: 10 respectively for 5 minutes.

3.4.2 Preparation of chemical bath

The composition of the chemical bath was maintained with 5 ml of 0.12 M :

CdCl2, 10 ml of 0.3 M : NH4Cl, 5 ml of 0.7M : CS(NH2)2, 30 ml of 2M : NH4OH and

50 ml of de-mineralized H2O. The bath temperature was maintained at 90 °C by a

constant voltage regulator. The substrate was suspended vertically inside the chemical

bath using teflon-disc substrate holder and was rotated at 20 rpm. Care was taken to

avoid any mechanical disturbance from the experimental set-up to the growth except

substrate rotation. All the experiments were conducted over a period of one hour by

51
keeping the chemical bath stationary. After each deposition the coated substrate was

cleaned in de – mineralized H2O and dried.

3.4.3 Results and discussion

3.4.3 (a) Structural properties

Cadmium sulfide has been reported to exist in two crystalline modifications viz.

cubic and hexagonal close packed (wurtzite 2H) structure. Both these structures are

closely related and share several common interplanar spacing ‘d’ with periodic stack

of cadmium atoms that are tetragonally co-oriented to sulfur atoms. The free energies

of both these structural modifications differ marginally and hence the coexistence of a

mixed plane of CdS has often been observed during CdS growth [60]. The

diffractograms of chemical bath deposited CdS thin film on different substrates are

shown in figure 18. The particle size ‘D’ was determined using the following

Scherrer’s formula [61].

D = [0.9 λ / β1/2 cosθ] nm (2)

where β1/2 is the full width at half maximum (FWHM) of XRD peak, θ is diffraction

angle and λ is X-ray wavelength used (1.5408 Å, Cu Kα). The strain component (ε)

was calculated using the value of β1/2 and the Bragg’s angle (θ) as explained in the

following relation [62].

ε = [(β1/2 cosθ) / 4] (3)

The dislocation density (ρ) is defined as the dislocation lines per unit volume of

crystal and can be evaluated from the Williamson and Smallman method [63-64],

ρ = (1/D2) lines/m2. (4)

52
Figure 18 The XRD spectra of the CdS film on (a) glass, (b) ITO coated
glass, (c) chemically polished p-Si <111> and (d) chemically
polished p-Si <100> substrates.

53
 Figure 19 illustrate the calculated crystallite size D and the corresponding ε and ρ

values of each peak in a spectrum (or for a sample). Notations (1) and (2) in the figure

19, represents the graphs for D versus ε and D versus ρ respectively. The average

crystallite size (Davg), the average strain (εavg) and the average dislocation density

(ρavg) for a sample was calculated using the D values and their corresponding ε and ρ

values of all the peaks in a XRD spectrum. The estimated structural parameters were

shown in the table 2. In general, figure 19 reveals that the gradual increment of the

crystallite size reduces the number of particles per volume thereby decrement in the

strain value. Data in table 2, infers that the choice of substrate influences the

crystallite size, strain and dislocation density. The magnitude of the dislocation

density observed in this study is in line with the reported values for thin films

comprising nanaoparticles [65-67].

Table 2 The average crystallite size (Davg), strain (εavg) and dislocation density
(ρavg) of CdS deposited on glass, ITO and p-Si substrates.

Substrate Average Average strain Average dislocation density

crystallite size (εavg x 10-4) (ρavg x 1014 lines/m2)
(Davg)
nm
Glass 37 7 7.3

ITO 19 22 2.7
p-Si<111> 33 9 9
p-Si<100> 18 24 4.7

54
Figure 19 Graphs of crystallite size ‘D’ versus strain ‘ε’ (1) and crystallite
size ‘D’ versus dislocation density ‘ρ’ (2) for CdS films on (a)
glass, (b) ITO-coated glass (c) chemically polished p-Si <111>
and (d) chemically polished p-Si <100> substrates.

55
3.4.3 (b) Formulation of 3-dimensional crystal shapes

The crystal shapes of the deposited nanocrystalline CdS film on different

substrates were computed using the crystal shape algorithm as published earlier [68].

With the use of 2θ value and the peak width at half-maximum (i.e., β) value for the

recorded X-ray diffraction peaks, the crystallite thicknesses, thkl, along the recorded

directions are computed using the Scherrer’s formula. The set of crystal planes

defined by the set of computed thkls’ is used to determine the mathematical envelope

for the crystal shape. These mathematical envelopes give the three-dimensional crystal

shapes. An elegant mathematical technique has been developed and tested for

computing the shapes of nanocrystals from X-ray diffraction (XRD) data. The efficacy

of this method for the shape analysis of crystallites in nanopowder specimens was

already demonstrated by comparing the computed shape and the shape visualized

using transmission electron microscopy (TEM). Unlike visualization techniques such

as TEM, the reported algorithm can formulate the three dimensional crystallite shapes

even in the presence of agglomeration in the film which may acquire any shape as

dictated by the experimental conditions during the film formation. In principle, the

said method was found to be applicable to powder XRD as well as single – crystal

XRD of nanosized as well as micro-sized particles. By applying this algorithm, the

computed crystal shapes for the film deposited in the present work are presented in

figure 20.

56
 Figure 20 The computed crystal shapes of nano CdS on (a) glass, (b) ITO
coated glass, (c) chemically polished p-Si <111> and (d)
chemically polished p-Si <100> substrates.

The chemical bath deposited CdS thin films on glass, ITO coated glass, chemically

polished p-Si <111> and chemically polished p-Si <100> substrates exhibit different

crystal shapes even though the experimental conditions were deliberately kept same.

The exposed Miller indices (hkl) and the exposed area (computed from the crystal

shape algorithm) of the three representative samples are presented in table 3.

Table 3 The calculated crystal area for the deposited CdS crystallites on
different substrates.

Substrate Faces indexed Crystal area (Å2)

Glass (002) 19914

(200) 8565
(220) 6875
ITO (002) 19353
(020) 16729
(220) 13526
p-Si<111> (002) 19051
(101) 3078
(110) 15179
(200) 5904
p-Si<100> (111) 4695
(002) 19838
(200) 8308
(220) 7850
(311) 7850

57
 The CdS film on glass, ITO coated glass and polished p-Si <111> and <100>

substrates show a disordered cubic crystal shape with well defined faces, a nearly

cubic crystal shape, a disordered hexagonal and a nearly hexagonal crystal shape with

well defined faces respectively. The observed influence on crystal shapes may be

attributed to the self-influence of the chosen substrates and their preparation

conditions since the nucleation mechanism of crystallites on the substrates are very

crucial to analyze the crystal shapes.

3.4.3 (c) Surface morphological properties

Figure 21 shows the SEM images of chemical bath deposited CdS thin films on

glass, ITO coated glass, chemically polished p-Si <111> and p-Si <100> substrates

deposited at 90° C. All the obtained micrographs revealed a well packed, more

uniform smooth surface of CdS without grown-in defects such as pinholes, cracks and

island growth. This observation suggested that the near optimized experimental

conditions prevailed during the deposition even though the substrates were different.

The SEM images show that the deposited films are constituted by the stacking of CdS

crystallites, which results from the coalescence of nano – grains and some larger sized

granules. In the present work, the observed aggregates are spherical, irrespective of

the substrate but the computed crystal shapes estimated from the XRD data differs

with respect to substrates. The crystal shape algorithm computes the shape of the tiny

crystallites present in the agglomerates, while the SEM shows only shape of the

agglomerates. The observed spherical features as seen in SEM (figure 21) are actually

agglomerates as the result of CdS nuclei formed and dispersed in the growth solution.

58
Figure 21 The SEM images of the CdS on (a) glass, (b) ITO coated glass,
(c) chemically polished p-Si <111> and (d) chemically polished
p-Si <100> substrates.

59
At a certain size range, these nuclei tend to form colloidal CdS spheres (either on

substrate or in solution), then it would spontaneously aggregate locally with their

neighbors to form clusters, each of which containing a few to a few tens of the CdS

nanoparticles [45]. But, the crystal shapes shown in figure 20 corresponds to tiny

crystallites which make up these spherical agglomerates. It should be further noted

that the spheres are micro sized while the computed crystallites are nanometer sized.

The mean size of the observed granules are in the range of 0.3 µm to 0.5 µm for CdS

deposited on glass, ITO coated glass and p-Si substrates. As a first attempt to

investigate the crystals shapes in the formation of thin films by chemical bath

deposition, more work is needed to establish a good correlation between the

computation and the experimental results. Such works are in progress. Depending

upon the kinetics of the synthesis route; the material may have equilibrium shape or

non-equilibrium shape. The employed algorithm basically works with the XRD

pattern of the synthesized powder recorded at room temperature and utilizes the well-

known Scherrer’s formula to compute the crystal shape; hence the information

obtained from this study may be utilized as an additional tool to tune the material

towards a specific application.

The elemental compositions of chemical bath deposited CdS thin films on three

different substrates such as glass, ITO-coated glass and p-Si were found to be having

the S / Cd ratio of 0.93, 0.94 and 0.97 respectively by energy dispersive X-ray

analysis.

60
3.4.3 (d) Photoluminescence properties

Figure 22 depicts the recorded fluorescence spectra with an excitation wavelength

of 330 nm on CdS films formed on glass, ITO coated glass, p-Si <111> and p-Si

<100>. In general, the role of surface quality, flatness and native defects of the

deposited material is very important in terms of emission characteristics towards

optical application. Also, it is well known that the photoluminescence in II-VI

semiconductors is influenced by the stoichiometric defects at electronic levels. In the

present PL studies with an excitation wavelength of 330 nm on the chemical bath

deposited CdS on the three different samples viz., CdS/glass, CdS/ITO coated glass

and CdS/p-Si <111> exhibit a low energy band around 2.34 eV and a high energy

band around 2.55 eV which resembles the characteristic PL spectrum at 10 K [69]. In

literature, the observed yellow (2.14 eV) and red (1.8 eV) emissions for CdS are

associated with the electronic transitions involving the interstitial-cadmium and

sulfur-vacancies, respectively. The observed green (2.25 eV to 2.4 eV) band might

originate from a transition involving a surface defect and valance band or the

conduction band and interstitial sulfur. For chemical bath deposited CdS, usually as-

grown samples do not show luminescence at room temperature, because of the high

density of non-radiative defects with energy levels near the midgap. But the recorded

spectrum with intense peaks at room temperature evidences the crystalline and optical

quality of the film. According to Baccarani et al. [70], the sulfur enrichment in

chemical bath deposited CdS films cause a reduction in the carrier trap density at the

grain boundaries, which accounts for low number of non-radiative centers giving rise

to an enhancement of the PL signal at room temperature. In the present work, the

obtained S / Cd ratio in the film showed near stoichiometry even though the starting

61
composition in the chemical bath had excess sulfur component. The substantially

enhanced photoluminescence from the CdS film deposited on silicon <100> substrate

as compared to other samples indicates that the silicon <100> substrate has more

influence in enhancing the surface quality and flatness of the film as well as reducing

the native defects in the deposited CdS film.

Figure 22 The room temperature PL emission spectra recorded for CdS thin
film on (a) glass, (b) ITO coated glass, (c) chemically polished p-
Si <111> and (d) chemically polished p-Si <100> substrates
under the excitation wavelength of 330 nm.

62
3.5 Effect of EDTA on CdS thin films

3.5.1 Introduction

Ethylenediamine tetraacetic acid (EDTA) is an amino - derived organic -

compound known to be a strong hexdentate chelating agent. Molecular formula of

EDTA is C10H16N2O8. It forms a complex with metal ions present in the solution and

dissociates reversibly at a low rate [71]. Application of EDTA in the chemical bath

solution to enhance the structural, compositional, optical and room temperature PL

emission properties of chemical bath deposited CdS film was investigated. The

observed beneficial effect on the crystallization, optical transparency, surface

morphology and photoluminescence emission of CdS crystallites in the bath solution

followed by deposition on ultrasonically cleaned glass substrates were elaborated.

3.5.2 Preparation of chemical bath

Chemical bath deposition of CdS thin films on glass substrates were carried out

from pure and EDTA – added chemical bath. In each experiment, the chemical bath

conditions were kept identical except the molar concentration of EDTA. The EDTA

solutions were prepared in four different concentrations such as 0.003 M, 0.004 M,

0.005 M and 0.006 M, respectively. The glass substrates of size, 1 x 1 cm2 were

etched for 5 minutes in 1% hydro fluoric (HF) acid solution to improve the adherence

of the film. Initially, the chemical bath was prepared with 7.5 ml of 0.2 M : CdCl2,

20ml of 0.2 M : NH4Cl, 7.5 ml of 0.4 M : CS (NH2)2, 20ml of 2 M : NH4OH, 60 ml of

de-mineralized H2O and 20ml of EDTA solution having pre-determined

concentration. The pH of the bath was maintained at 9 by tuning the concentration of

NH4Cl concentration in the chemical bath.

63
 The deposition of CdS thin films from the said aqueous solution containing EDTA,

the following reaction would happen.

Cd2+ + EDTA → Cd(EDTA)2+

Cd(EDTA)2+ + 2e- → Cd + EDTA

The bath temperature was maintained at 90° C. The substrates were suspended

vertically inside the chemical bath using teflon disc (1 x 1 cm2) substrate holder and

rotated with an rpm of 20. Care was taken to avoid any mechanical disturbance from

the experimental set-up to the deposition except substrate rotation mechanism. All the

experiments were conducted over a period of 60 minutes by keeping the chemical bath

stationary. After each deposition the coated substrates were cleaned in de-mineralized

H2O and dried in air atmosphere. The structural analysis was carried out through

powder X-ray diffractometer system. The SEM and EDX were done to investigate the

surface morphology as well as the stoichiometry of the grown CdS films. The room

temperature PL emissions on all the samples were recorded through PL emission

spectroscopy. All the deposited CdS films are observed as golden yellow in color and

the granules are uniformly distributed over the substrate, highly adherent and

specularly reflective.

3.5.3 Results and discussion

3.5.3 (a) Structural and compositional properties

The X-ray diffraction spectra of the deposited CdS thin films from pure and

EDTA – added solution having different concentrations are shown in figure 23. The

observed major reflections are identified as per the standard JCPDS card nos., 80-

0006 (hexagonal – H), 89 – 0440 (cubic – C) and 80 – 0019 (hexagonal- H). In the

case of CdS deposited from pure solution reflects hexagonal structure whereas the

64
CdS film deposited from EDTA – added solution possess crystallites belongs to

hexagonal and cubic crystal structures. On comparison with the CdS film deposited

from pure solution, the increasing concentration of EDTA in the solution reduces the

amorphousity and enhances the crystallinity of the film. The EDTA concentration of

0.006 M has resulted better crystalline quality than all other studied samples. The co-

existence of both hexagonal and cubic structured crystals in the deposited film led us

to investigate further to obtain a film having either hexagonal or cubic structure by

optimizing the EDTA concentration in the bath solution.

Figure 23 The X-ray diffraction spectra recorded on the CdS thin films
deposited from pure and EDTA added bath solution.

65
 Compositional analyses on the deposited CdS films were done by EDAX and are

displayed in figure 24. The calculated S/Cd ratio for the CdS films deposited from

pure and EDTA – added solutions having concentrations of 0.004, 0.005 and 0.006 M

are as follows; 52 / 48, 51 / 49, 49 / 51, and 49.41 / 50.59. This shows that the

enhancement in the concentration of EDTA in the chemical bath solution has

beneficial effect in controlling the stoichiometry of the film.

Figure 24 The EDAX analysis on the CdS thin films deposited from (a)
pure and EDTA –added bath solutions having concentrations of
(b) 0.004 M, (c) 0.005 M and (d) 0.006 M.

66
3.5.3 (b) Surface morphology and texture properties

Figure 25 illustrates the SEM micrographs on the deposited CdS thin films from

pure and EDTA-added bath solution. A close observation on these pictures revealed a

fact that the addition of EDTA in the bath solution greatly reduces the grain size and

macroscopic defects such as voids and pinholes in the thin film. Further, the observed

increase in the number of grains per area (~ 30 / 1µm2) suggested that as the

concentration of EDTA increases in the bath solution, it enhances the formation of

CdS crystallites in the bath solution followed by deposition on the glass substrate. The

morphology of the grains is observed to be near spherical in nature and is more

uniform throughout the analyzed area in the specimen.

Surface features of the deposited CdS films were characterized by atomic force

microscopic analysis. Four different EDTA concentrations in the bath solution exhibit

different surface texture properties as shown in the figure 26. The surface texture

parameters such as root mean square surface roughness, average height and maximum

height have been calculated by using the previously reported AFM software [72]. The

measured surface texture parameters are presented in the table 4. The addition of

EDTA in the bath solution influences the surface roughness of the deposited film.

67
Figure 25 The surface micrographs of the CdS thin films deposited from
EDTA –added bath solutions having concentrations of (a) 0.003
M, (b) 0.004 M, (c) 0.005 M and (d) 0.006 M.

68
Figure 26 The AFM analysis on the CdS thin films deposited from EDTA –
added bath solution having concentrations of (a) 0.003 M, (b)
0.004 M, (c) 0.005 M and (d) 0.006 M.

69
 Table 4 The calculated surface texture parameters by using the AFM
software.

EDTA Root mean square Average height Maximum height

concentration surface roughness (µm) (µm)
(mol / L) (µm)
0.003 0.037 0.15 0.4

0.004 0.036 0.16 0.39

0.005 0.032 0.08 0.26

0.006 0.027 0.073 0.18

3.5.3 (c) Optical properties

Optical transmittance curves in the wavelength range of 200 - 1100 nm were

recorded on the CdS thin films deposited from pure and EDTA-added bath solutions

and are depicted in the figure 27. From the recorded spectrum, one can infer that the

CdS film deposited from EDTA - added solution shows higher % of transmittance

when compared to the film deposited from pure solution. The film deposited from the

bath solution containing the EDTA concentration of 0.006 M attains 80% of

transmittance over other films. And, all the deposited films are transparent in the

wavelength range of 550 - 1100 nm and showed a narrow cut-off around 420 nm.

From this study, it is concluded that the addition of EDTA in the bath solution also

enhances the optical quality of the film by reducing the grown-in defects during the

crystallization and macroscopic defects such as pinholes and voids during the

deposition.

70
 Figure 27 The optical transmittance spectra of the CdS thin films.

3.5.3 (d) Photoluminescence properties

The room temperature photoluminescence (PL) emission spectra of the chemical

bath deposited CdS thin films from pure and EDTA-added bath solution are recorded

using the excitation wavelength of 390 nm and are presented in the figure 28. The

present study indicated that the intensity of the CdS-characteristic emissions around

470 and 512 nm depends on the added EDTA concentration in the bath solution.

Also, a noticeable shift towards green emission was observed while increasing the

concentration of the EDTA in the bath solution. In the case of film deposited from

0.006 M EDTA added solution, there was no blue emission and only green emission

around 504 and 550 nm. The actual emission wavelengths are given in the following

table 5.

71
 Figure 28 Recorded room temperature PL emission spectra of the deposited
CdS thin films.

Table 5 Observed characteristic PL emission wavelength of the deposited CdS

thin films.

EDTA concentration in the bath Observed emission wavelength (nm)

solution (mol / L)

0.00 473 and 512

0.003 485 and 529

0.004 490 and 535

0.005 495 and 542

0.006 504 and 550

72
3.6 I-V characteristics of the fabricated n-CdS/p-Si <100> heterojunction

CdS thin films were deposited on to polished p-Si <100> substrates under

predetermined bath condition (with out EDTA). The fabricated heterojunction have

been characterized for I-V characteristics. Commercially bought silver paste has been

realized for the formation of ohmic-contact for p-Si <100> and n-CdS. A typical I - V

characteristic curve of n-CdS/p-Si <100> heterojunction was shown in the figure 29.

Ideality factor of the fabricated heterojunction was calculated by using the formula,

Ideality factor n = {(q / KB T) [(V2 ~ V1) / ln (I2 / I1)]}

Where q is the charge of an electron and q = 1.6 x 10-19 C

KB is the Boltzmans constant and T be the temperature in Kelvin

V2 and V1 are the maximum and minimum applied voltage

I2 and I1 are the corresponding current flowing through the fabricated diode

The value of (q / KB T) is approximately 26 x 10-3. The fabricated polished n-CdS

/p-Si <100> heterostructure showed good rectification characteristics and the

calculated ideality factor was ~ 1.3. The estimated ideality factor of the fabricated n-

CdS /p-Si <100> heterostructure is in line with the previously reported values [73].

73
Figure 29 The current - voltage characteristics of CdS / polished p-Si <100>
heterojunction.

74
3.7 Conclusions

In conclusion, the chemical bath deposition of CdS thin films on glass, ITO coated

glass and chemically polished p-Si <111> and p-Si <100> substrates have been

carried out using an in-house fabricated experimental set-up. The efficacy of the

fabricated experimental set-up towards the realization of good quality films are

justified by means of various characterization studies involving structural, optical and

electrical properties of the thin films. The crystallographic structure, crystallite size

and strain were investigated by the X-ray diffraction. The average crystallite sizes

calculated from the extracted X-ray diffraction data are 37, 19, 33 and 18 nm for CdS

on glass, ITO coated glass, p-Si <111> and p-Si <100> substrates respectively. The 3

dimensional shape of the CdS nanocrystals grown on the substrates has been

visualized with the help of an algorithm which explicit that the crystal shape / habit

changes due to the selection of substrates even though the experimental parameters are

deliberately kept identical. It was found that the nanocrystalline CdS on glass

substrate exhibits the disordered cubic crystal shape and the film on ITO coated glass

substrate showed a habit transformation from disordered cubic to near cubic crystal

shape. The CdS film on chemically polished p-Si <111> substrate was found to have a

hexagonal crystal shape with less disorderliness where as film on p-Si <100> resulted

near perfect hexagonal shape. The effect of experimental parameters on the surface

morphology of the CdS films on all the four representative substrates were

investigated with the help of scanning electron microscopic analysis. It was found to

be closely packed granules with less pinholes for the film on p-Si when compared to

other two representatives. The mean sizes of these granules are observed in the range

of 0.3 - 0.5 µm for CdS deposited on glass, ITO coated glass and p-Si substrates. The

75
identical surface features exhibited on all the samples confirms the near optimized

experimental conditions irrespective of the different substrates used. The PL spectrum

of nano CdS exhibits a low energy band around 2.34 eV and a high energy band

around 2.55 eV which resembles the characteristics of the reported recorded at 10 K-

PL spectrum. Further, the characteristic of the PL intensity changes with the shape of

the crystallites. The present investigation revealed the tendency of CdS crystallites to

adopt a shape based on the selection of the substrate. When compared with other two

studied substrates, the degree of organization, crystallinity, crystal habit,

stoichiometry and the PL intensity was better in the case of CdS film on p-Si <100>.

One of the well studied chelating agent viz., EDTA was added in the chemical bath

solution to investigate the effect of its concentration during the chemical bath

deposition of CdS thin films. Chemical bath deposited CdS thin films were deposited

on glass substrates from pure and EDTA - added bath solutions having different

concentrations of 0.003, 0.004, 0.005 and 0.006 M. The observed beneficial effects on

the physical and optical properties of the deposited CdS thin films has strongly

suggested that the addition of EDTA in the bath solution modifies the chemical

kinetics of the chemical bath through its chelating action thereby the crystalline

quality of the CdS crystallites was improved much. The optical transparency of the

film attains a maximum value of around 80% over the wavelength range of 650-1100

nm for the film deposited from EDTA - added solution having concentration of 0.006

M. Also, the observed intensity of the characteristic room temperature PL emission

was found to be influenced by the concentration of EDTA in chemical bath solution

during the formation of CdS thin films. A through investigation on the kinetics of its

76
chelating action will throw more light on the actual mechanism prevails during the

deposition and may helps to obtain a film with desired properties.

The current – voltage characteristic of the fabricated n-CdS/p-Si<100>

heterojunction was investigated through I-V measurements. The current - voltage

curves for the fabricated heterojunction traced under forward and reverse bias

conditions showed a near ideal diode characteristics having the ideality factor of ~ 1.3.

77
3.8 References

[1] Egon Wiberg, Arnold Frederick Holleman, Inorganic Chemistry, Elsevier

(2001).

[2] Traill R J, Boyle R W, American Mineralogist 40 (1955) 555.

[3] D. Lincot, Gary Hodes, Proceedings of the international symposium, The

Electrochem. Soc. (2006).

[4] Greenwood, N. Norman, A. Earnshaw, Chemistry of the Elements, Oxford:

Butterworth-Heinemann (1997).

[5] P. H. P. Wanrooij, U. S.Agarwal, J. Meuldijk, Kasteren, J. M. N. Van, P. J.

Lemstra, J. Appl. Polymer Sci. 100 (2006) 1024.

[6] Mario Schiavello, Photoelectrochemistry, Photocatalysis, and Photoreactors:

Fundamentals and Developments; Springer (1985).

[7] A.F. Wells, Structural Inorganic Chemistry; 5th edition, Oxford Science

Publications (1984).

[8] Antonio Luque, Steven Hegedus, Handbook of Photovoltaic Science and

Engineering, John Wiley and Sons (2003).

[9] D. Reynolds, G. Leies, L. Antes, R. Marburger, Physical Review 96 (1954)

533.

[10] C. Fouassier, Luminescence in encyclopedia of inorganic chemistry, John

Wiley & Sons (1994).

[11] Minkus, Wilfred, Physical Review 138 (1965) A1277.

[12] Smith, Roland, Physical Review 105 (1957) 900.

[13] Yu A. Akimov, et al. Soviet J. Quantum Electron. 2 (1972) 284.

78
[14] Agarwal, Ritesh; Barrelet, J. Carl, Lieber, M. Charles, Nano Letters 5 (5)

(2005) 917.

[15] G. Sasikala, R. Dhanasekaran, and C. Subramanian, Thin Solid Films 302

(1997) 71.

[16] G.A. Kitaev, A.A. Uritskaya, and S.G. Morkrushin, Russian J. Phys.

Chem. 39 (1965) 1101.

[17] I. Kaur, D.K. Pandya, and K.L. Chopra, J. Electrochem. Soc. 127

(1980) 943.

[18] T.L. Chu, S.S. Chu, N. Schultz, C. Wang, and C.Q. Wu, J. Electrochem.

Soc. 139 (1992) 2443.

[19] J. Choi, K.-J. Kim, J.-B. Yoo, and D. Kim, Solar Energy 64 (1998) 41.

[20] M.A. Martinez, C. Guillen, and J. Herrenro, Appl. Surf. Sci. 140 (1999) 182.

[21] G. Brunthaler, M. Lang, A. Forstner, C. Giftge, D. Schikora, S. Fereira,

H. Sitter, and K. Lischka, J. Crystal Growth, 138 (1994) 559.

[22] H.C. Chou and A.R. Ohatgi, J. Electronic Materials 23 (1994) 31.

[23] H.C. Chou and A.R. Ohatgi, J. Electrochem. Soc. 142 (1992) 303.

[24] T.L. Chu, J. Britt, C. Ferekides, C. Wang, and C.Q. Wu, IEEE, Transactions on

Electron Devices Letters 13 (1992) 303.

[25] Y.F. Nicolau, Appl. Surf. Sci. 22/23 (1985) 1061.

[26] S.A. Al Kuhaimi, Vacuum 51 (1998) 349.

[27] R.W. Birkmire, B.E. McCandless, and S.S. Hegedus, Solar Energy,

12 (1992) 45.

[28] Jae-Hyeong Lee, Journal of Electroceramics 17 (2006) 1103.

79
[29] Jae-Hyeong Leea, Woo-Chang Songa, Jun-Sin Yia, Yeong-Sik Yoob, Solar

Energy Mater. Solar Cells 75 (2003) 227.

[30] J. Zhan, X. Yang, D. Wang, S. Li, Y. Xie, Y. Xia, Y. Qian, Adv. Mater. 12

(2000) 1348.

[31] Y. Li, H. Li, Y. Ding, Y. Qian, L. Yang, G. Zhou, Chem. Mater. 10 (1998)

2301.

[32] S. Yu, Y. Wu, J. Yang, Z. Han, Y. Xie, Y. Qian, X. Liu, Chem. Mater. 10

(1998) 2309.

[33] J. Huang, Y. Xie, B. Li, Y. Liu, Y. Qian, S. Zhang, Adv. Mater. 12 (2000) 808.

[34] Y. Xie, J. Huang, B. Li, Y. Liu, Y. Qian, Adv. Mater.12 (2000) 1523.

[35] Hui Zhang, Deren Yang, Xiangyang Ma, Duanlin Que, Mater. Lett. 59 (2005)

3037.

[36] H. Zhang, X.Y. Ma, Y.J. Ji, J. Xu, D.R. Yang, Chem. Phys. Lett. 377 (2003)

654.

[37] H. Zhang, D.R. Yang, Y.J. Ji, X.Y. Ma, S.Z. Li, D.L. Que, Mat. Chem. Phys.

93 (2005) 65.

[38] Hui Zhang, Deren Yang , Xiangyang Ma, Materials Letters 61 (2007) 3507.

[39] Ran Zhai, ShuBo Wang, HaiYan Xu, Hao WangT, Hui Yan, Mat. Lett. 59

(2005) 1497.

[40] K.J. Rao, B. Vaidhyanathan, M. Ganguli, P.A. Ramakrishnan, Chem.

Mater. 11 (1999) 882.

[41] Yonghong Ni, Xiang Ma, Jianming Hong, Zheng Xu, Materials Letters 58

(2004) 2754.

80
[42] Qingchun Zhaoa, Wenming Chena, Qingren Zhu, Materials Letters 57 (2003)

1756.

[43] J.S. Fawcett, C.J.O.R. Morris, Separation Science 1 (1966) 9.

[44] K. Yu. Ponomareva, I. D. Kosobudsky, E. V. Tret’yachenko, and G. Yu.

Yurkov, Inorganic Mater. 43 (2007) 1160.

[45] Xin-Yong Tao, Xiao-Bin Zhang, Ji-Peng Cheng, Wan-Zhen Huang, Materials

Letters 59 (2005) 2689.

[46] R Devi, P Purkayastha, P K Kalita, and B K Sarma, Bull. Mater. Sci., 30

(2007) 123.

[47] Wang Qingqing, Zhao Gaoling, Han Gaorong, Mater. Lett. 59 (2005) 2625.

[48] A. V. Feitosa, M. A. R. Miranda, J. M. Sasaki, and M. A. Araujo-Silva,

Brazilian J. Phys. 34 (2004) 656.

[49] S. J. Lade, and C. D. Lokhande, Mat. Chem. Phys. 49 (1997) 160.

[50] J.W. Park, K. R. Min, and E. Y. Jeung, B. Kor. Chem. Soc. 9 (1988) 408.

[51] T. I. Igumenova et al., J. Photochem. and Photobiology A: Chem. 94 (1996)

205.

[52] T. Sugimoto, G. E. Dirige, and A. Muramatsu, J. Colloid and Interface Sci. 182

(1996) 444.

[53] D. M. Zhou, S. Q. Wang, and H. M. Chen, J. Environ. Sci. Chi. 13 (2001) 153.

[54] R. Padmavathy et al., Mat. Lett. 53 (2002) 321.

[55] Jae-Hyeong Lee, J. Electroceram. 17 (2006) 1103.

[56] R. Orlega-Borges, D. Lincol, J. Electrochem. Soc. 140 (1993) 3464.

[57] M. A. Martinez, C. Guillen, J. Herrero, Appl. Surf. Sci. 136 (1998) 8.

81
[58] P. Mendoza-Perez, G. Santana-Rodriguez, J. Sastre-Hernandez, A. Morales-

Acevedo, Thin Solid Films 480 (2005) 173.

[59] X. Wu, J.C. Keane, R.G. Dhare, A. Duda, T.A. Gessert, P. Sheldon, 17th

European photovoltaic solar energy conf. Munich, Germany, October 22-26,

2001, p. 995.

[60] F.C. Phillips, An Introduction to Crystallography; The University Press:

Glasgow, Scotland, 1971.

[61] Kaur, D.K. Pandya, K.L. Chopra, J. Electrochem. Soc. 127 (1980) 943.

[62] I. Kaur, D.K. Pandya, and K.L. Chopra, J. Electrochem. Soc. 127 (1980) 943.

[63] R. Bhattacharya, A. Manna, T. K. Das, and S. Saha, Cryst. Res. Technol. 45

(2010) 656.

[64] S.Venkatachalam, D.Mangalraj, K.Narayandas, K.Kim, and J.Yi, Physica B

358 (2005) 27.

[65] Zhao, D.L.Morel, and C.S. Ferekides, Thin Solid Films 413 (2002) 203.

[66] K. Ravichandran, B. Sakthivel, and P. Philominathan, Cryst. Res. Technol. 45

(2010) 292.

[67] R. Das and R. Kumar, Cryst. Res. Technol. 45 (2010) 725.

[68] D. Sherwood and B. Emmanuel, Cryst. Growth Des. 6 (2006) 1415.

[69] J. Aguilar-Hernandez, J. Sastre-Hernandez, R. Mendoza-Perez, M. Cardenas-

Garcia, and G. Contreras-Puente, Phys. Status Solidi (c) 2 (2005) 3710.

[70] G. Baccarani, B. Ricco, and G. Spadini, J. Appl. Phys. 49 (1978) 5565.

[71] Mudi Xin, KunWei Li, Hao Wang, Applied Surface Science 15 (2009)1436.

[72] I.Hprcas, R.Fernandez, J.M.Gomez-Rodriguez, J.Colchero, J.Gomez-Herrero

and A.M.Baro. Rev. Sci. Instrum, 78 (1) (2007) 013705.

82
[73] M.G.Mahesha, V.Kasturi, G.Bangera, K.Shivakumar, Mater. Sci.

Semiconductor Proces. 12 (2009) 89.

83