Influence of growth time on the properties of CdTe thin films grown by electrodeposition using acetate precursor for solar energy application

Cadmium telluride (CdTe) thin films were deposited using a two–electrode electrodeposition (ED) configuration from an aqueous acidic solution. The electrolyte solution contains 1 M of cadmium acetate dihydrate (Cd (CH3OO) 2.2H2O) as cadmium precursor and 1 ml of tellurium dioxide (TeO2) as tellurium precursor. The thin films were grown for different deposition times of 60, 120, 180, 240, and 300 min to investigate the effect of the deposition period on the structural, optical, electrical, surface morphology, elemental composition, and surface roughness properties of the CdTe thin films in both as–deposited and heat–treated forms. X-ray diffraction (XRD) analysis indicates that the CdTe thin films have polycrystalline cubic zinc blend, orthorhombic and hexagonal structures. The result confirmed that the cubic phase is dominant and the peak for preferred orientation is along the (111) plane. Ultraviolet-visible (UV–vis) spectrophotometry study shows that the band gap of the as-deposited thin films varies from (1.41–1.45) eV, and after heat treatment, the band gap decreased to (1.39–1.42) eV. Photoelectrochemical cell (PEC) measurements show that CdTe thin films haven-type conductivity in both as–deposited and annealed forms. Scanning electron microscopy (SEM) analysis shows that the surface morphology of CdTe thin films changed as the deposition period increases. After heat treatment, increase in grain size was observed. Energy–dispersive x-ray spectroscopy (EDS) analysis shows that the percentage composition of as–deposited and heat-treated CdTe thin films varied with deposition time. After post–deposition treatment (PDT), the concentration of Te decreased, while that of Cd increased due to recrystallization during annealing. For the film deposited for 120 min, stoichiometric composition of CdTe was observed after heat treatment. Scanning probe microscopy (SPM) measurements revealed that the average surface roughness of the thin films varied with deposition time. The maximum average surface roughness was recorded when the film was deposited for 120 min. These results show that the prepared CdTe thin films have potential application as absorber layers in thin film solar cells.


Introduction
Due to their high absorption coefficient and direct energy band gap in the range (1.45-3.7) eV, semiconductor thin films of II-VI compounds, such as CdTe, CdSe, CdS, ZnS, and ZnSe are used for solar cell applications [1,2]. Specifically, CdTe thin film is used as absorber material for solar energy harvesting in solar cells owing to its near-ideal band gap of about 1.5 eV [3]. Solar energy is one of the best renewable and sustainable energy sources compared to others such as wind, ocean/tide, and hydropower. Due to its availability and huge energy supply, it has attracted the attention of researchers over the years. However, the main challenges using solar energy technologies include low efficiency, high cost of production, environmental friendliness, and stability issues affecting the solar cell's performance. Besides the efficiency of the cell, for the affordability of this device, production cost reduction is the central research area in solar energy conversion. The solution to this problem is using a thin film solar cell with a low-cost growth method [4].
The CdS/CdTe solar cell has been recognized as one of the potential possibilities for developing sound photovoltaic systems [5]. CdTe has recently gained attention due to its many applications, such as photovoltaics [6], x-ray sensors [7], solar energy water splitting [8], imaging detectors [9], optoelectronics [10], and radiation detectors [11].
The structure, morphology, surface roughness, and optical properties of CdTe thin films depend on the growth method and optimization of growth parameters [5]. There are several methods to grow CdTe thin films, such as vacuum evaporation [12], chemical molecular beam deposition (CMBD) method [13], close space sublimation (CSS) [14], physical vapour deposition [15], successive ionic layer adsorption and reaction (SILAR) [16], pulsed laser deposition (PLD) [17], metal-organic chemical vapour deposition (MOCVD) [18], coprecipitation [19], Radio frequency (RF) sputtering [20], sol-gel [21], chemical bath deposition (CBD) [22] and electrodeposition (ED) [23]. Each growth technique has its advantages and disadvantages. When a thin film is grown by MOCVD and physical vapour deposition (PVD) methods, the phases move immediately from gas to solid. However, phase shifts from liquid to solid occur during electrodeposition. As nature favours this transition, it should result in layers of high-quality materials.
Electrodeposition as a thin film deposition technique has comparative advantages concerning deposition process continuity, simplicity, low cost, scalability, self-purification, easy doping, manufacturability, bandgap engineering ability, and Cd-containing waste reduction [24].
There are different sources of Cd for the electrodeposition of CdTe thin films, such as cadmium chloride [26], and cadmium sulphate (CdSO 4 ) [27]. The structure, morphology, and optical properties of these films are governed by optimization of deposition parameters such as precursor concentrations and ratios of precursor concentrations used to prepare the bath electrolyte, pH of the solution, deposition temperature, cathodic deposition voltage, rate of stirring of the electrolyte, electrodes used, and deposition period.
For solar cell device applications, the absorbance of the layer depends on the film thickness which is significantly controlled by deposition time [28]. In the case of CdTe, its thin film has a high absorption coefficient, and to absorb up to 99% of the visible part of solar spectra, about 2 μm thickness is required [27]. In the development of high-efficiency CdTe-based thin film solar cells, the reduction of the thickness of the CdS window/buffer layer for minimizing absorption loss, the modification of conductive glass fluorine doped tin oxide (FTO), indium doped tin oxide (ITO), etc. Post-deposition CdCl 2 treatment, and using two-electrode configuration instead of conventional three-electrode are significant steps that have been taken for the current status of high performance of the device [29]. Electrodeposition from an aqueous bath is usually performed at a low temperature of less than 90°C due to the boiling point of water being 100°C.
For the CdTe thin film growth, two-electrode and three-electrode configurations are used in the electrodeposition system [1,30]. In this paper, we use a two-electrode method since it has some advantages compared to three electrodes. The advantages include reducing the cost of production by eliminating the reference electrode, increasing the deposition temperature up to 90°C (since reference electrodes cannot operate safely above 70°C), reducing the impurity metal ions from the reference electrode in the case of accidental leakage or breakage, reducing the growth time as a result of increased growth temperature, and overall process simplification [26]. Therefore, in the present work, we have used two-electrode systems.
Electrodeposition of CdTe thin film from an ionic liquid bath containing a combination of cadmium acetate and 1-butyl-3-methylimidazolium acetate as Cd precursors and using three-electrode cells was reported by Waldiya et al [31]. To the best of our knowledge, the use of cadmium acetate as the Cd source for the cathodic electrodeposition of CdTe thin films from an aqueous solution using a two-electrode electrodeposition method, has not been reported.
The main aim of this study, therefore, is to investigate the effect deposition time on the structural, optical, electrical, surface morphology, elemental composition, and surface roughness properties of electrodeposited CdTe thin films on glass/FTO by using two-electrode configuration and with cadmium acetate dihydrate as the only source of Cd.

Materials and method
CdTe thin films were grown from an aqueous solution containing 1.0 M cadmium acetate dihydrate [Cd(CH 3 OO) 2 .2H 2 O] (98% purity) in 400 ml of de-ionized water, as cadmium precursor and 1 ml of tellurium dioxide (TeO 2 ) solution (99% purity) as tellurium precursor. Both chemicals were laboratory reagent grade purchased from Sigma Aldrich.
The thin films were deposited using two electrode electrodeposition method coated glass/FTO substrate with a sheet resistance of 8 /square. Before starting the electrodeposition process, the glass/ FTO substrates were cut into 2.5 × 2.5 cm 2 and washed ultrasonically with a laboratory soap solution in deionized water for 30 min, followed by ethanol, acetone, and methanol, respectively. The samples were rinsed in deionized water (DI) upon washing and finally dried in air. The electrolyte solution pH was adjusted to 2.00 using dilute HCl or ammonium hydroxide (NH 4 OH) at room temperature. The deposition temperature of the electrolyte solution bath was maintained at 85°C, and during deposition, the electrolyte solution was stirred at a moderate stirring rate using a magnetic stirrer. The source of electrical power used for the two-electrode system was a computerized Gill AC potentiostat (ACM instrument, United Kingdom). An insulating polytetrafluoroethylene (PTFE) thread seal tape was used to attach the glass/FTO substrate to a high-purity graphite rod which was used as the working electrode (cathode), and the graphite rod as a counter electrode (anode). Before CdTe thin film deposition takes place, the cadmium acetate solution was electro-purified for 48 h at a deposition voltage less than that of element Cd, which is determined by cyclic voltammetry study. Since the purity of the Cd source used in this study is low, so the impurity metal ions need to be removed from the electrolyte solution.
The Te-containing solution was prepared by dissolving 2 g of TeO 2 in 30 ml of dilute HCl solution and then stirring for two hours since Te is insoluble in water and soluble in an acidic medium. 1 ml of this TeO 2 solution, was added to the 400 ml of Cd-containing solution, and then the mixture was stirred for 300 min to make it homogeneous. The addition of HCl was selected for pH reduction in this study. In addition to pH adjustment, we used it as a source of Cl for n-type doping of the CdTe thin films. During the electrodeposition of CdTe, the deposition voltage was maintained at 1250 mV, while deposition time was 60, 120, 180, 240, and 300 min, respectively. After the deposition of CdTe thin films at different times, the films were cut into two equal halves. A set of one-half was left as-deposited (AD), and the other set (of the second half) was dipped into CdCl 2 solution for 5 s and then dried in air. For this post-deposition treatment, the CdCl 2 -containing solution was prepared as 0.1 M CdCl 2 in 30 ml DI water in a 50 ml beaker. This second set was further annealed at 400°C for 20 min in air. The structural, optical, electrical, surface morphology, elemental composition, and surface roughness properties of the samples were investigated in order to establish the best deposition time. The experimental setup for two electrode electrodeposition method is shown in figure 1. The reference electrode was removed for cost minimization, impurity reduction, and overall process simplification in the thin film synthesis process.

Cyclic voltammetry (CV) study
To determine the reduction potential of ions in an electrolyte solution cyclic voltammetry (CV) is used. It is an experimental technique that can examine the reduction and oxidation processes in an electrochemical system [32]. The CV measurement was performed using a glass/FTO electrode in the aqueous acidic solution containing 1.0 M of Cd (CH 3 OO) 2 .2H 2 O and 1 ml of TeO 2 , with a graphite counter electrode. During CV measurement, the solution temperature was maintained at 85°C. The solution's pH was initially adjusted to 2.00 at ambient temperature, and the CV measurement was performed using a cathode potential voltage range of 0-2000 mV at a cyclic sweep rate of 10 mV s −1 . The element Te has a higher positive standard reduction potential (E 0 = +0.593 V) compared to Cd (E 0 = −0.403 V) with respect to the standard hydrogen electrode [25].
As shown in figure 2, one observes that Te starts to deposit at a cathodic potential of about 390 mV in the forward direction at point (A). When the deposition voltage increases, the element Cd begins to deposit at a cathode potential of about 945 mV. At this point (B), the co-deposition of Cd and Te to form CdTe is initiated. 1.0 M cadmium acetate dihydrate Cd (CH 3 OO) 2 . 2H 2 O and 1 ml of TeO 2 ion sources.
The film formed in this voltage region is Te-rich CdTe. Further increase in deposition voltage showed that the current density increased near (1100−1500) mV in the area (C) where the formation of more stoichiometric CdTe thin film was growing in this region. When the deposition voltage was at 1600 mV, region (D), there was an increase in current density for the growth of CdTe with Cd-richness. Further increase in deposition voltage leads to a high increase of current density due to cadmium dendrite formation or electrolysis of water or both [33].
In the reverse direction in figure 2, the negative current density shows the removal of CdTe from the glass substrate or Te from the cathode. For this study, the deposition voltage was kept at 1250 mV, and we investigated the effect of the deposition period on the quality of the film.
The following reactions can be used to deposit CdTe films on the cathode: The overall reaction is

Structural properties
The structure and crystallinity of the electrodeposited CdTe thin films were determined by utilizing a Bruker D8 Advance x-ray diffractometer with a monochromatic wavelength of 1.5416 Å. The x-ray generator's voltage and current were kept constant at 40 kV and 40 mA, respectively. To determine the crystal structure of the CdTe layers formed on the glass/FTO, the incident angle was adjusted to a range of 20 to 70°.   (107), respectively. The film deposited at 60 min, produced mixed phases of cubic, orthorhombic, and hexagonal structures. With a further increase in the deposition time to 120 min, the structure was mainly cubic with high intensity of (111) preferential orientation peak, and a better structure was observed. When the film was grown for 180, 240, and 300 min, mixed phases of cubic, orthorhombic, and hexagonal were observed again. These results agree with the Joint Committee on Powder Diffraction and Standards (JCPDS) reference file numbers 752086, 410941, and 800090 for cubic, orthorhombic, and hexagonal structures, respectively.  015), and (107) planes, respectively. When the deposition time was 60 min, the (111) peak was increased. This shows that during the annealing process, the free Te atoms reacted with Cd, and recrystallization took place. These results agree with energy dispersive x-ray spectroscopy (EDS) result. Further, an increase in the deposition time to 120 min produced film with a polycrystalline cubic structure with intensity of the (111) preferential orientation peak decreasing and the other two peaks increasing in intensity, which agrees with the other reported results [34]. The mixed phase structures of polycrystalline cubic, orthorhombic, and hexagonal CdTe were also observed for the samples deposited for 180, 240, and 300 min.
The crystallite sizes of the ED-CdTe thin films were calculated using Scherer's formula in equation (4) [35].
Where the parameter D is the crystallite size, β is the full width at half maximum (FWHM)of the (111) peak under consideration, in radian, λ is the Wavelength of the x-rays (0.15406 nm for Cu Kα), and θ is the Bragg's diffraction angle in degree.
The dislocation density δ, which specifies the length of dislocation lines per unit volume of crystal in the thin-film CdTe, was computed by using equation (5) [36].  Figure 4 shows the crystallite size for as-deposited CdTe as a function of deposition time. The crystallite sizes calculated were 19, 41, 30, and 30 nm for deposition times of 60, 120, 240, and 300 min, respectively. The maximum crystallite size was at 120 min deposition time.
For solar cell application, the thickness of CdTe thin film absorber materials is a crucial factor, and in electrodeposition, this thickness depends on growth time. The film thicknesses were determined theoretically by using Faraday's law (equation (6)) of electrolysis [37] and experimentally by scanning electron microscopy (SEM) cross-section measurement.   Where J is the average current density during deposition, t is the deposition period, m is the molar mass of CdTe (240.01 g.mol −1 ), n is the number of electrons transferred for deposition of 1 molecule of CdTe (n = 6), F is the Faraday constant 96,485 C.mol −1 , ρ is the density of CdTe, and T is the film thickness.
Both theoretical and experimental results show that as deposition time increased, the thickness of the film increased. For further analysis, the samples prepared at deposition times of 60, 120, and 240 min, were selected for characterization. A thickness of 1.75 μm was recorded at 120 min deposition time and 3.5 μm at 240 min.
As shown in figure 5, the theoretical thickness of CdTe is higher than the experimental values. The reason for the observed difference is that in the electrodeposition process, due to the electrolysis of water at a significantly high potential, all electrons do not participate in film formation. However, in general, the film thickness increases as deposition time increases within the range of time under consideration.

Optical properties
Optical absorbance measurements were conducted on as-deposited and heat-treated CdTe thin films. Measurements were performed using ultraviolet-visible (UV-VIS) spectrophotometer win lab-scan lambda 950 in order to determine the energy band gap of the films. Figure 6(a) shows the graph of absorbance as a function of wavelength for the as-deposited films. It was observed that the absorbance of CdTe thin films varied with deposition time for both as-deposited and heattreated samples.
The absorbance of the as-deposited film increased with deposition time. This shows optimization of deposition time has a significant impact on the absorbance of the film. Figure 6(b) shows that the absorbance increased after annealing, resulting from thermally induced re-crystallization and a total reaction between unreacted Cd and Te atoms to form CdTe. This enhancement in absorbance after heat treatment reveals that annealing plays an important role in improving the properties of CdTe thin films, and this is vital for solar energy applications.
The optical band gap of electrodeposited CdTe (ED-CdTe) was estimated from the absorbance squared (A 2 ) as a function of the photon energy plot. The energy band gap was determined by extrapolating the straight line portion of the graph to the photon energy axis at (A 2 = 0). As shown in figure 6(c), the energy band gaps of the asdeposited films were recorded as (1.41-1.45) eV for the corresponding deposition times. Figure 6(d) shows the graphs of square of absorbance versus photon energy for estimating the band gap of the annealed ED-CdTe thin film samples. The band gap values were found to decrease and were recorded as (1.39-1.42) eV for deposition times of (60-300) min. This bandgap change is attributed to the removal of some impurities and pinholes in the film in the annealing process. The impurities, in this case, are not additional elements in the films, but internal impurities related to elemental composition variation in Cd and Te in CdTe thin film, such as excess un-reacted Te as well as Cd vacancies, which are well known to occur in CdTe [26,33,34,38]. At 120 min of growth time, after heat treatment, the band gap of the prepared film is close to that of bulk CdTe, which is 1.41 eV. This report agrees well with the previously reported CdTe by using the electrodeposition method with different sources of Cd reported by Dharmadasa et al [39].

Photoelectrochemical cell analysis (PEC)
Photoelectrochemical cell (PEC) measurements were carried out to estimate the electrical conductivity types of the ED-CdTe layers. This experimental technique typically investigates the potential barrier formation at the solid/liquid interface between CdTe and a suitable electrolyte solution when glass/FTO/CdTe is immersed in the electrolyte as a cathode and graphite rod as an anode, which is connected to a digital voltmeter to measure the potential when the circuit is under dark and illuminated conditions. The PEC result is the difference between the voltages under illumination and dark conditions representing the PEC signal of the solid/liquid junction as reported by Echendu et al [40].
The sign of the PEC signal for the particular system used is negative for n-type conductivity and positive for p-type conductivity [34]. For the PEC measurements, the prepared electrolyte solution containing 0.1 M Na 2 S 2 O 3 in 20 ml of deionized water was stirred for 120 min to make it homogeneous. Figure 7(a) shows that PEC signals of the CdTe thin films varied with deposition time. For both the as-deposited and annealed samples, the PEC signals were negative values. This shows that the CdTe thin films were n-type in conductivity. It is known that CdTe conductivity is determined by the concentration of Cd and Te within the film. The conductivity is p-type if CdTe thin film is Te-rich and n-type if it is Cd-rich [41]. In this study, we used HCl for pH adjustments and as a source of Cl dopant to grow n-type CdTe for n-CdS/n-CdTe device fabrication. Reports in the literature show that the use of n-CdTe absorber in solar cell fabrication helps to obtain highefficiency cells with high short-circuit current densities. The comparative study reported by Echendu et al on n-CdS/p-CdTe and n-CdS/n-CdTe analysis confirmed that n-CdS/n-CdTe device structure has less recombination, with increased short-circuit current density and therefore higher efficiency [42]. The work by Schulmeyer et al also supports the achievement of higher cell efficiency with n-CdTe [43]. This structure is also favourable for back metal contact fabrication, which is a cause for back surface recombination [42].

Morphological property
Scanning electron microscopy images were taken to investigate the surface morphology, grain size, and uniform coverage of the glass/FTO substrate by CdTe. For this analysis, the films deposited for 60, 120, and 240 min were selected. Figure 8 shows that the substrate surface was covered uniformly, while grain sizes and shapes varied with deposition time. When the film was deposited at 60 min, the grain sizes were small, and after heat treatment, there was an increase in grain size. As the deposition time was increased to 120 min, the grains became larger and more uniform. The grain sizes increased when the samples were heat treated, and much clear morphology was observed at this deposition time. Further increase in the deposition time to 240 min caused the grains to decrease in size, and gradual peeling of thin films from the glass/FTO substrate was observed.
The grain size modification and large grains were visible after heat treatment. These large grains have significant importance for solar cell device applications. The overall improvement in grain size after heat treatment was due to the recrystallization of grains which is in good agreement with the XRD results, and as the deposition time increased, the thickness of the films also increased, as shown in figure 5. The results were confirmed, as shown in figure 8 for the cross-section of the SEM images that increasing deposition time resulted in an increase in thickness.

Elemental composition
The elemental composition of as-deposited and heat-treated CdTe thin films was investigated using energydispersive X-ray spectroscopy measurements. For this analysis also, the films deposited for 60, 120 and 240 min are selected. As shown in figure 9, the expected elements Cd and Te are present in the EDS spectra, and the elemental composition varies with respect to deposition time. When the film was as-deposited for 60, 120, and 240 min, there was a change in the Cd and Te percentage composition. Table 1 indicates that for 60 min deposition time, the amount of Te is more in as-deposited film and reduced after heat treatment. When the film was deposited for 120 min, the as-deposited film was still Te-rich but closer to stoichiometric with 45% of Cd and 55% of Te.
Further increase in the deposition time to 240 min showed that the as-deposited film is still Te-rich since Cd was removed from the surface when the deposition time was long in an acidic medium. This situation happens in electrodeposition. After heat treatment, the composition of Cd was modified in the CdTe films, with the films becoming nearly stoichiometric for all the deposition durations. From the results, one sees that the film grown for 120 min shows the best stoichiometry in both as-deposited and heat-treated forms. For the film that was grown for 120 min, stoichiometric composition (50, 50) was observed after heat treatment.
The Si, F, Sn, and O peaks observed in the EDS spectra are from the glass/FTO substrate, and the Cl element is either from the CdCl 2 used for chemical treatment before heat treatment or from the HCl used for

Scanning probe microscopy (SPM)
The surface roughness of ED-CdTe thin films was investigated using scanning probe microscopy (SPM). For solar energy applications, studying surface properties is essential. Especially for absorber materials, the absorbance depends on the surface roughness of the thin film. The average surface roughness of the samples presented in the present work varied with deposition time. Again, only the films deposited for 60, 120, and 240 min were selected for this characterization.  surface roughness was increased to 89, 134, and 73 nm, respectively. The maximum average surface roughness was recorded for the film grown for 120 min, and heat treatment enhanced the deposited film's roughness.

Conclusion
CdTe thin films were successfully deposited by using a two-electrode electrodeposition configuration from an aqueous bath containing cadmium acetate as Cd source and tellurium oxide as Te source. The study shows that changing deposition time impacts the properties of the film. XRD analysis reveals that the thin films were polycrystalline cubic structures at 120 m deposition time for both as-deposited and heat-treated films. Cubic, orthorhombic, and hexagonal structures were observed when the film was grown at 60 and 240 min. UV-VIS spectroscopy measurements show that the absorbance of CdTe thin film varied with deposition time in both asdeposited and heat-treated samples. The energy band gap of as-deposited CdTe varies from 1.41 to 1.45 eV, and after heat treatment in the presence of CdCl 2 , the bandgap varies from 1.39 to 1.42 eV with deposition time. PEC measurement confirmed that the conductivity of CdTe thin film was n-type in both as-deposited and heattreated samples. SEM images show that the CdTe thin films covered the surface of the substrate with increased grain sizes after heat treatment. EDS analysis shows that the elemental composition at 60 min film shows Terichness. The film was near stoichiometric composition and had a better Cd/Te ratio when the deposition time was 120 min. In all the films, the heat treatment improved the composition of Cd in the film, and the film became more stoichiometric. SPM images show that the average surface roughness varies with growth time. After post-deposition heat treatment, there was an improvement in the surface roughness. The average surface roughness was better at 120 min deposition time for both as-deposited and heat-treated samples. Since the film's surface properties significantly contribute to the absorbance layer, optimizing deposition time and postdeposition heat treatment with CdCl 2 was very important for solar energy application.