Analysis of electrodeposited CdTe thin films grown using cadmium chloride precursor for applications in solar cells

CdTe thin films were electrodeposited cathodically on glass/FTO substrates by potentiostatic technique in which the anode was a high purity graphite rod. Cadmium chloride hydrate CdCl₂·xH₂O of 99.995% purity was utilised as cadmium precursor, while, tellurium oxide TeO₂ of 5 N (99.999%) purity was used as the tellurium source. The electrolyte was prepared to make 1.5 M CdCl₂·xH₂O in 400 ml deionised (DI) water contained in a 500 ml polypropylene beaker. The polypropylene beaker was placed inside a 1000 ml glass beaker containing DI water. The glass beaker serves as the outer DI water jacket which helps in maintaining uniform heating of the electrolyte in the polypropylene beaker. Prior to the addition of TeO₂, the CdCl₂·xH₂O aqueous solution was electro-purified for ~100 h by applying a cathodic potential just below the required potential for reduction of Cd²⁺. Afterwards, low level of TeO₂ solution was added to the solution and stirred for ~5 h to achieve solution homogeneity. It should be noted that TeO₂ is partially soluble in some acidic media but insoluble in water [14]. Therefore, 0.03 M of TeO₂ solution was prepared by dissolving ~2 g of TeO₂ in concentrated 30 ml hydrochloric acid. The acidic media was stirred for ~1 h to achieve homogeneity before the TeO₂ is gently diluted with 400 ml of DI water in a polypropylene conical flask. All these steps were taken before TeO₂ addition in minute quantity into the electro-purified CdCl₂·xH₂O electrolytic bath. An initial volume of 5 ml TeO₂ solution was added to the CdCl₂·xH₂O electrolytic bath to give a total of 0.0023 M TeO₂ concentration. The bath was continuously stirred and the temperature was maintained at ~85 °C. It should be noted that the Te concentration in the electrolytic bath was kept comparatively low due to its high positive reduction potential of Te as compared to Cd.

The pH at the start of the deposition was adjusted to 2.00 ± 0.02 at room temperature using diluted solutions of hydrochloric (HCl) and ammonium hydroxide (NH₄OH). It should be noted that the 85 °C bath temperature used during the growth of CdTe layer was due to higher crystallinity achievable at higher growth temperature [7], but due to the aqueous solution utilised in this experiment, the growth temperature is limited. In this work, two-electrode configuration was utilised. This comprises of a high purity carbon anode and a 3 × 4 cm² glass/FTO cathode with a sheet resistance of 7 Ω/sq.

The power supply source was a computerized Gill AC potentiostat. Prior to the deposition of CdTe layer, cyclic voltammetric test was performed to determine the cathodic potential range in which stoichiometric or near stoichiometric CdTe can be grown. It should be noted that, immediately after CdTe growth, the glass/FTO/CdTe was divided into two halves. One-half was left as-deposited while the other was CdCl₂ treated (CCT) at 400 °C for 20 min in air. Experimentation on just heat treated CdTe layers were not performed due to improved material and electronic properties observed after CdCl₂ treatment [1]. The CdCl₂ treatment was carried out by dipping the as-grown CdTe layer into 0.1 M CdCl₂ aqueous solution for 2 s, retracted and allowed to dry in air before heat treatment. The treatment with CdCl₂ was necessitated due to its material and device quality improvement as reported by numerous independent researchers in the literature. All the substrates and chemicals used in this study were procured from Sigma-Aldrich Ltd, UK.

The glass/FTO substrates were cleaned using ultrasonic bath containing soap solution for ~15 min. The substrates were rinsed thoroughly afterwards in DI water, dried in a stream of nitrogen gas before it is been degreased using acetone and methanol. The substrates were immediately rinsed in DI water and attached to the cathode rod using polytetrafluoroethylene tape. Finally, the glass/FTO substrates were rinsed in DI water and transferred directly into the electrodeposition bath.

Cyclic voltammogram is the current–voltage characteristic of an electrolyte which provides information on the elemental/compound deposition taking place at different deposition potentials. The study helps in identifying the range in which stoichiometric or near-stoichiometric CdTe layers can be achieved. The level of crystallinity, crystalline structure and phase identification of the as-deposited and post-growth treated CdTe layers were obtained using X-ray diffraction (XRD) technique. This information was extracted using Philips PW 3710 X’pert diffractometer with Cu-Kα monochromator of wavelength λ = 1.54 Å. The X-ray generator tension was set to 40 kV while the current was adjusted to 40 mA for this set of experiments. Optical property such as the absorbance was studied using Carry 50 Scan UV–Vis spectrophotometer at room temperature between the wavelengths of 200–1000 nm. A thoroughly cleaned TEC7 glass/FTO substrate was setup as the baseline. Raman spectroscopy studies were performed on samples using Renishaw InVia Raman spectrometer using a 514 nm argon ion laser excitation source to determine the crystallinity, identify the phases present and also to determine the phonon modes of both the as-deposited and the CdCl₂ treated samples. The Raman spectroscopy laser power and objective were set to 50% and ×100 respectively. The thickness of the CdTe layers was measured using UBM Microfocus optical depth profilometer (UBM, Messtechnik GmbH, Ettlingen, Germany). The surface morphology and compositional analysis of both the as-deposited and CdCl₂ treated CdTe films were studied using FEI Nova 200 NanoSEM equipment. The electrical conductivity type of the CdTe layers was determined using photoelectrochemical (PEC) cell measurement as an alternative to the conventional Hall effect measurements. The Hall effect measurement of CdTe deposited on FTO are not possible due to the underlying conducting FTO substrate.

PEC measurement was conducted by dipping the glass/FTO/CdTe layer into an aqueous electrolyte containing 0.1 M of sodium thiosulphate to form a solid/liquid junction. The difference between the voltages measured under white light illuminated (V _L ) and dark (V _D ) conditions provides the open circuit voltage or the PEC signal. The sign of the PEC signal determines the electrical conductivity type of the CdTe layer [15].

The DC conductivity measurements on glass/FTO/CdTe/Ohmic contact structures were carried out on the electrodeposited CdTe layers to determine the electrical resistivity (ρ) and conductivity (σ) of the layers.

Figure 1 shows the cyclic voltammogram of an aqueous solution containing a mixture of 1.5 M CdCl₂·xH₂O and a low level of pre-prepared TeO₂ solution in 400 ml of DI water during the forward and reverse cycle between 100 and −2000 mV cathodic voltage. The electrolytic bath pH was adjusted to 2.00 ± 0.02 using dilute solutions of HCl and NH₄OH. The stirring rate and bath temperature were maintained at 300 rpm and ~85 °C respectively. Tellurium deposits first in the electrodeposition process due to its standard reduction potential value of +593 mV with respect to standard H₂ electrode which is more positive than that of cadmium with standard reduction potential value of −403 mV.

It was observed from Fig. 1 that the deposition of Te starts in the forward cycle at cathodic potential of ~400 mV under the experimental conditions used (inset in Fig. 1) according to the following electrochemical reaction:

With an increase in cathodic potential, more Te is deposited. The deposition of Cd and formation of CdTe starts around ~1000 mV cathodic potential as depicted by the first hump shown in Fig. 1. Tellurium-rich CdTe is expected at the initial stage of deposition of cadmium, afterward, with an increase in deposition potential and there exists a narrow cathodic potential window in which stoichiometric CdTe compound can be deposited. Further increase in the cathodic potential above this point results in an increase in the cadmium richness of electrodeposited CdTe. The electrochemical equation for Cd deposition and the complete CdTe formation is given in Eq. (2) and (3) respectively.

It should be noted that the steep increase in current density (marked c) at a cathodic voltage above ~1400 mV might be due to either the formation of Cd dendrites on the working electrode or the electrolysis of water at any deposition potential above ~1230 mV [14]. Both scenarios have been reported to have a detrimental effect on the quality of the deposited CdTe layer [16]. However, the release of most active hydrogen atoms on the material surface also introduces an advantage of hydrogen-passivation during growth although the formation of hydrogen bubble could be a disadvantage due to a possible cause of material layer peeling. Therefore, selecting the growth voltage has a crucial effect on the growth of suitable CdTe layer. In the reverse cycle, the dissolution of elemental Cd and Cd from CdTe occurs in the voltage range of ~1500 and ~1000 mV, while the dissolution of Te from the glass/FTO substrate occurs below 750 mV as shown in Fig. 1.

Therefore, based on the information obtained from the voltammetric study, cathodic potential range between 1000 and 1400 mV was pre-characterised using XRD for as-deposited CdTe layers grown at a step size of 50 mV (results not presented in this paper). The highest peak intensity signalling highest crystallinity was observed at 1350 mV, therefore, surrounding cathodic potentials were scanned at 10 mV intervals and characterised to identify the best cathodic potential in which stoichiometric and near-stoichiometric CdTe can be achieved.

The DC resistivity experiment was performed on CdTe layers grown between the cathodic voltage of 1330 and 1400 mV. The CdTe layers were grown for 120 min each. 2 mm diameter of 100 nm thick gold contacts were evaporated at a high pressure of 10^− 5 Nm⁻² on the p-type CdTe layer for both the as-deposited and CdCl₂ treated CdTe layers (glass/FTO/p-CdTe/Au), while 100 nm thick indium contacts were evaporated at a high pressure of 10⁻⁵ Nm⁻² on the n-CdTe layers for both the as-deposited and CdCl₂ treated CdTe layers (glass/FTO/n-CdTe/In) to achieve Ohmic contacts for the metal/semiconductor interfaces [28, 29]. The electrical resistivity (ρ) of the layers was calculated using equation (v) where R is the electrical resistance, A is the contact area and L is the film thickness. The average electrical resistance (R) was calculated using the I–V data extracted under dark condition using Rera Solution PV simulation system.

Figure 10 shows the plot of electrical resistivity (ρ) against the cathodic voltage in which the CdTe layers were grown. From observation in Fig. 10, the resistivity of CdTe layers reduces after CdCl₂ treatment which might be due to defect and grain boundary passivation, reduction of grain boundaries due to grain growth during CdCl₂ treatment, increase in CdTe crystallinity amongst other advantages of CdCl₂ treatment as documented in the literature.

For both the as-deposited and CdCl₂ treated CdTe layers grown at ~1360 mV cathodic voltage exhibit the high resistivity value. This is expected since the CdTe is stoichiometric and intrinsic in electrical conduction. The material grown at 1360 mV should, therefore, have the highest resistivity. However, cathodic voltages closer to the intrinsic voltage (V _i ) would be favourable for the fabrication of devices due to high crystallinity, better optical and morphological properties achieved in stoichiometric or near-stoichiometric CdTe layers. It should be noted that based on the understanding as published by Dharmadasa et al. (2002) [21, 30, 31] and new results as published by independent researchers such as Reese et al. (2015) [23] and Burst et al. (2016) [22], fabricating devices with slightly Cd-rich CdTe is favourable due to increased carrier lifetime, defect reduction amongst other advantages.