Electrochemically synthesized microcrystalline tin sulphide thin films: high dielectric stability with lower relaxation time and efficient photochemical and photoelectrochemical properties

Abstract

A detailed study has been carried out on the structural, dielectric and impedance properties of polycrystalline p-type SnS thin films grown on transparent conducting oxide (TCO) coated glass substrates from an aqueous solution of tartaric acid, SnSO₄ and Na₂S₂O₃ by a modified electrochemical technique. The as-deposited films were found to be smooth, almost pinhole free and well adherent to the bottom substrate. X-ray diffraction studies revealed the formation of polycrystalline SnS films with an orthorhombic phase. Field emission scanning electron microscopy and atomic force microscopy revealed a moderately compact surface morphology with evenly distributed almost spherical grains. Optical measurements showed direct band gap energy of 1.5 eV. Detailed electrical (dc and ac) analyses showed the p-type nature of the deposited films with unique dielectric behavior. The band-gap energy, resistivity, dielectric constant and relaxation time make this material and ideal absorber layer, which is also reflected in the efficient photochemical and photoelectrochemical behavior.

---

1. Introduction

In recent years, thin films of SnS (an IV–VI semiconductor) have attracted much attention for photovoltaic applications due to their high absorption coefficient (≈10⁴ cm⁻¹ near the fundamental edge)^1,2 and high conductivity (mobility ∼90 cm² V⁻¹ s⁻¹). The constituent elements are inexpensive, nontoxic and abundant in nature, leading to the development of devices that are environmentally safe and have public acceptability. SnS is an important optoelectronic material, normally found in the orthorhombic phase (lattice constant: a = 0.385 nm, b = 1.142 nm and c = 0.438 nm), which is also called herzenbergite.^3,4 According to previous reports, the electronic transition in SnS is mostly direct in nature with the room temperature band gap energy of 1.2–1.5 eV for the bulk system. These properties enable SnS thin films for probable use as the absorber layer in the fabrication of hetero-junction solar cells.^5,6

SnS thin films can be fabricated by various methods like thermal evaporation,⁶ vacuum evaporation,^2,7 spray pyrolysis,⁸ chemical bath deposition,⁹ electrochemical deposition,¹⁰ RF sputtering,¹¹ SILAR method,¹² pulsed electrodeposition,¹³ electron beam evaporation,¹⁴ to name a few. For all these characteristics and relatively convenient deposition methodologies, SnS seems to be quite tailor-made as an absorber layer for thin film solar cells,¹⁵ photoelectrochemical solar cells¹⁶ and photocatalysts.^17,18 Apart from photovoltaic conversion; it has potential applications in optical data storage systems¹⁹ and solar control devices²⁰ as well. Electrochemical deposition technique provides some key advantages compared to other fabrication techniques, like good control over deposition parameters and film thickness, minimum wastage of raw materials, low cost, selective area deposition, etc.

In this work, we report a simplified electrochemical route, called the galvanic route for the deposition of SnS thin films on SnO₂:F coated transparent conducting oxide (TCO) glass substrate, which does not involve any external energy source like the conventional electrochemical technique, for carrying out the deposition. The films were found to have high dielectric stability with lesser relaxation time which leads to efficient photochemical and photoelectrochemical properties.

2. Experimental

2.1. Deposition procedure

To start with, the TCO substrates were cleaned with detergent and then dipped into concentrated chromic acid solution for about 30 minutes and washed thoroughly with cold distilled water to remove any adhering impurities. They were then boiled in methanol and after drying, were degreased in a vapor of trichloroethylene.

The working solution was prepared by mixing thoroughly 0.214 g of SnSO₄ and 0.075 g of tartaric acid in 100 ml double distilled water. Then 1.24 g of solid Na₂S₂O₃ was added to it and gently stirred (500 rpm) with magnetic stirrer for 15 minutes. The pH of the working solution was maintained around 2.5 by adding dilute sulphuric acid (1 : 10) drop wise. The working solution was transparent.

To carry out the deposition, a properly cleaned TCO glass substrate and a Zn rod were dipped into above working solution. The Zn rod and the TCO glass substrate were short-circuited externally through a copper wire. The Zn rod served as a self-decaying anode and the TCO glass as the cathode (schematic is reported elsewhere).²¹ As soon as the two electrodes were short-circuited, the deposition starts, which was continued for 60 minutes at 27 °C without stirring, to obtain compact films with good adherence and almost without any pinholes. After deposition, the films were washed thoroughly with double distilled water and dried in a hot air oven at 50 °C for 15 minutes.

2.2. Characterization

The crystalline structure and phase of the deposited films were determined by X-ray diffraction (XRD) technique using SEIFERT 3000 P X-ray diffractometer with Bragg–Brentano goniometer geometry and Cu-K_α X-radiation source (λ = 1.540598 Å). Morphological analyses of the films were carried out by both energy field emission scanning electron microscopy (FESEM) (JEOL, JSM-6700F) and atomic force microscopy (AFM) (NT-MDT SolverPro). The thickness of the films was determined gravimetrically using a Mettler Toledo AB-54-S balance. UV-vis spectroscopic measurements of the films for evaluation of bandgap energy and for photocatalysis were carried out using a JASCO V-530 UV-vis spectrophotometer. Two-probe resistivity and photoelectrochemical cell performances were measured using a KEITHLEY-4200 semiconductor characterization system. A Phillips made 200 W/240 V tungsten filament lamp was used for illuminating the dye solution for photocatalysis and the SnS electrode for photoelectrochemical measurements. The light intensity on the SnS surface was 100 mW cm⁻². The four-probe resistivity measurements were carried out using DFP-RM four-probe set up (SES Instruments, India). The dielectric properties of the films were analyzed using an Agilent 4284A Precision LCR meter. The Hall measurements were carried out using an ECOPIA HMS-5000 Hall Effect Measurement System (Korea) at a fixed magnetic field of 0.555 Tesla.

3. Results and discussions

3.1. Deposition chemistry

The mechanism for SnS deposition involves two reactions that take place simultaneously on the TCO cathode surface, when the electrodes were short circuited externally as described in Section 2.1. When SnSO₄ is dissolved in water at low pH (ca. 2.5), it dissociates as follows:

SnSO₄ → Sn²⁺ + SO₄²⁻ (i)

Again, Na₂S₂O₃ dissociates in the aqueous medium to produce S₂O₃²⁻ ions, which act as the source of sulphur for the deposition of tin sulphide.

Na₂S₂O₃ → 2Na⁺ + S₂O₃²⁻ (ii)

The electrodeposition process involved the formation of SnS thin films, which included simultaneous deposition of tin and sulfur from the bath by using two electrode system where properly cleaned TCO coated glass substrates were used as the cathode electrodes, whereas, Zn-rod electrode were used as the anode, respectively. pH 2.5 and 30 °C were found to be optimum for the deposition.

The basic electrochemical reactions for the simultaneous co-deposition of Sn and S should be characterized by E_Sn = E_S where E_Sn and E_S are the equilibrium deposition potential of the elements Sn and S respectively.²² The mechanism of SnS formation involves two reactions that had been taken place simultaneously on cathodic (TCO) surface.

For cathodic deposition:

2Sn²⁺ + 4e⁻ → 2Sn⁰, E⁰ = −0.1375 V (iii)

S₂O₃²⁻ + 6H⁺ + 4e⁻ → 2S⁰ + 3H₂O, E⁰ = +0.60 V (iv)

Overall cathodic reaction may be cited as:

2Sn²⁺ + S₂O₃²⁻ + 6H⁺ + 8e⁻ → 2SnS + 3H₂O (v)

Here, the formation of SnS film follows the conventional cathodic reduction pathways, where the Sn²⁺ and S₂O₃²⁻ ions are reduced simultaneously on the cathode surface by taking up electrons to form the desired film. For stoichiometric SnS deposition, the overall cathodic reaction is an eight electron transfer process. The detailed electrochemical reactions are given in ESI-1.†

In presence of complexing agent tartaric acid, the activity of Sn²⁺ ions can be controlled due to the formation of chelate complex to slow down the reactions at the vicinity of cathode surface and improvement of film quality.²³

Sn²⁺ + [C₄O₆H₆] → [Sn(C₄O₆H₆)]²⁺ (vi)

Sn²⁺ will form 1 : 1 complex with organic tartrate ions.²⁴ Here, we have used Zn rod as the sacrificial anode and it was dipped in to a low pH (ca. 2.5) medium, it will dissociate in the following manner:

Zn → Zn²⁺ + 2e, E⁰ = +0.76 V (vii)

The Zn²⁺ ions will come to the solution and as the Zn anode and TCO cathodes are short circuited externally, the two electrons that were released on the Zn surface, will flow to the TCO cathode through the external path, which in turn, will reduce the Sn²⁺ and S₂O₃²⁻ ions present in the vicinity of the cathode.

The effective formal potential of the overall cell reaction is much more affected due to pH and complexation affect, in this circumstances Zn/Zn²⁺ system will act as the driving force for the above reaction. This is the basic principle of a Galvanic Cell, and for this reason, we designate the process as “galvanic”, where, no external source of energy was used to carry out the required reactions.

As the cathodic reactions are simultaneous and competitive, we can write the overall cathodic reaction as:

2[Sn(C₄O₆H₆)]²⁺ + S₂O₃²⁻ + 6H⁺ + 8e⁻ → 2SnS + 3H₂O + 2[C₄O₆H₆] (viii)

The tartrate ions released by Sn²⁺ ions during reduction, capture the free Zn²⁺ ions present in the solution, and thus restricting their deposition along with Sn and enhancing the purity of the deposited films.

3.2. Growth kinetics

In order to find the film growth kinetics, the thicknesses of the deposited films were plotted against the deposition time, which is shown in Fig. 1. The films were deposited at room temperature at different time, keeping all other parameters constant. The thicknesses of the films were determined by gravimetric method using the following formula:

t = (W₂ − W₁)/Ad

where, ‘t’ is the thickness of the film, ‘A’ is the area covered by the film, ‘d’ is the density of SnS (5.197 g cm⁻³), W₁ and W₂ are the weight of the TCO substrate before and after deposition, respectively. From Fig. 1, it is evident that the film thickness increases almost linearly with time up to 60 minutes and after that saturation in thickness takes place. The saturation thickness was estimated to be about 2.5 μm, which was achieved at a growth rate of 4.2 × 10⁻² μm min⁻¹. It can be inferred that, after 60 minutes, the [Sn(C₄O₆H₆)]²⁺ complex acted as the source of Sn²⁺ in the solution, got exhausted by reacting with the thiosulphate ions (as there is no continuous supply of them) leading to the formation of SnS. Thus, due to the lack of source ions, the reaction was stopped and no further increase in the film thickness was observed. We have taken such saturated films with a thickness of 2.5 μm for all characterization.

Fig. 1 Deposition kinetics for SnS films.

3.3. X-ray diffraction study

It can be seen from the XRD pattern (Fig. 2) that all the diffraction peaks are coincident with the standard diffraction pattern of orthorhombic SnS (JCPDS ID 39-0354). The diffraction peaks at 2θ = 30.70°, 31.56° and 31.89° can be indexed to (101), (111) and (040) diffraction planes of SnS, respectively (inset of Fig. 2). The diffraction peaks marked “*” correspond to the SnO₂ of the bottom TCO coated glass substrate (JCPDS ID 21-1250). It is also evident from the XRD pattern that the (211), (220) and (321) diffraction planes of SnO₂ are superimposing with the (151), (061) and (341) planes of SnS. The lattice plane index (hkl), interplanar distance d_hkl and lattice parameters have the following relationship for orthorhombic system:²⁵

1/d_hkl² = h²/a² + k²/b² + l²/c²

Fig. 2 X-ray diffraction pattern of a representative SnS film deposited on TCO coated glass substrate.

This equation has been applied to calculate the lattice constants and found to be a = 4.34 Å, b = 11.21 Å and c = 3.98 Å which are in close agreement with the standard value of a = 4.329 Å, b = 11.192 Å and c = 3.983 Å. An estimation of the crystallite size for the polycrystalline SnS was obtained from the broadening of the XRD peaks according to the Scherer's equation²⁶ D = 0.9λ/β cos θ and found to be around 30 nm [D is the crystallite size, λ is the wavelength of the Cu-K_α radiation used (λ = 1.5405 Å), β is the experimentally observed diffraction peak width at half maximum (FWHM), θ is the Bragg angle]. However, the overlapping of the two peaks viz. (111) and (040) of SnS (Fig. 2 inset) may introduce some inaccuracy in determining the FWHM and related cos θ, which in turn, may generate errors in calculating the crystal size using Scherer's equation. This technique may be considered for the rough estimation of crystallite size, which needs to be verified further by other techniques like transmission electron microscopy as shown in this case as ESI-2.† The sharp peaks with relatively narrow FWHM indicate high crystallinity. Absence of diffraction peaks from other probable elements like free Sn, S or Zn (from Zn rod) and compounds like SnS₂ or ZnS, indicates the purity of the deposited material.

For the fabrication of good quality thin film to be used in optical devices, it is necessary to reduce the surface roughness and dislocation density of the film. The value of the dislocation density (δ) which gives the number of the defects in the film was calculated from the average values of the crystallite size D by the relationship δ = 1/D² (ref. 27) and found to be significantly low (Table 1). A comparison between the FWHM, crystallite size (D) and dislocation density (δ) at different (hkl) of the SnS films is shown in Table 1.

Table 1 Comparison between FWHM, grain size and dislocation density

(hkl)	FWHM (radian)	D (Å)	δ (×10^2 nm)^−2
101	0.0064	225	0.001975
111	0.0046	314	0.001014
040	0.00197	295	0.001149

---

3.4. Morphological analyses by SEM and AFM

In order to study the morphology of the deposited material, FESEM analysis was carried out and the related micrographs are shown in Fig. 3. From the low magnification (×30 000) FESEM image (Fig. 3a) it is evident that the surface of the deposited SnS film is composed of uniformly distributed spherical grains. The grains are distinct and the grain boundaries are prominent, leading to a moderately compact surface morphology. To study the nature of an individual grain, higher magnification (×100 000) image was captured and found that, such a grain is composed of numerous crystallites with average diameter in the range 30–50 nm (encircled in Fig. 3b). This value of crystallite diameter matches well with what we have obtained from the XRD analysis. From both the lower and higher magnification images, it has been found that the grain diameter of the galvanically deposited SnS film falls within the range 400 to 500 nm. However, the average grain size as obtained from FESEM is somewhat larger than that calculated from XRD analysis, which is attributed to the formation of larger grains from smaller crystallites, which is required for spontaneous lowering of Gibb's free energy in the system (ESI-3†). As the value of film thickness is crucial for different electrical measurements, we have carried out cross-sectional FESEM analysis (Fig. 3c) with the deposited film to be sure about the order of film thickness. From Fig. 3c, it is evident that the thickness of the film was about 2.5 μm, which is in good agreement with the gravimetrically obtained value.

Fig. 3 FESEM images of the SnS film (a) lower magnification and (b) higher magnification.

AFM is another important tool to study the morphology of a film. The AFM measurements were carried out in contact mode with a resolution of 5 μm × 5 μm and pitch height 300 nm. The 2D AFM image (Fig. 4a) shows that the film was composed of uniformly distributed spherical grains and the TCO glass substrate was well covered by the deposition. The spherical nature and average grain diameter (∼400 nm) are in good agreement with the findings from FESEM analysis. From the grain size distribution curve (Fig. 4b) it has been found that the particles have high order of size distribution, which supports their uniformity. About 85% of the particles were found to have 400 nm diameters. So, the process reported here is quite capable of producing SnS films with narrow size distribution and high order of uniformity.

Fig. 4 (a) Two-dimensional AFM image of a representative SnS film (b) grain size distribution curve.

3.5. Optical (UV-vis) characterization

The UV-vis absorbance spectra (Fig. 5) obtained for SnS films shows an onset of absorption at around 800 nm, which arises due to the near band edge transition in the semiconductor and indicates band gap energy of about 1.5 eV. The band gap energy was evaluated from the absorption spectra using the following relation:

(αhν) ∞ (hν − E_g)ⁿ

where, E_g is the energy band gap, α is the absorption coefficient, n is a coefficient having the value 1/2 for allowed direct band-to-band transition. The absorption coefficient α can be evaluated from the relation:

α = (1/t) × ln(100/T)

where, ‘t’ is the thickness of the film and T is the percentage transmittance.

Fig. 5 Optical absorption spectrum of the SnS thin film (inset: Tauc plot showing the band gap energy).

The Tauc plot (plot of (αhν)² vs. hν) is shown as the inset of Fig. 5, from which the direct band gap of the deposited material was found to be about 1.5 eV, ideal for an absorber material in thin film hetero-junction solar cells. This result also supports that the SnS films deposited by galvanic technique has no SnS₂ as impurity, which have a higher band gap energy value in the range 2.1 to 3.5 eV.²⁸

3.6. Electrical properties

3.6.1. Current–voltage (I–V) characteristics by two-probe technique. Two-probe I–V measurement was carried out taking TCO coated glass substrates as the bottom contact and 2 mm × 2 mm Au (evaporated by e-beam technique) as the top contact. The measurements were performed in a magnetically and electrically sealed chamber, under dark. Fig. 6 represents the I–V plot for the film, which shows typical diode characteristics between TCO and SnS. TCO being an n-type semiconductor, it can be inferred that the SnS films deposited on TCO are p-type in nature. Zener breakdown occurs at a greater reverse bias (around −2 V), which indicates better p-type nature of the deposited SnS films and good diode characteristics of the TCO/SnS/Au structure. This might be due to the large grain size of the films deposited on TCO, giving rise to low leakage path. On the other hand, from the FESEM and AFM images (Fig. 3a and b and 4a) it is evident that the film deposited has less micro-pores along the c-axis, which is responsible for producing less leakage path.

Fig. 6 Representative two-probe I–V characteristics for the SnS thin film under dark.

3.6.2. Current–voltage (I–V) characteristics by four-probe technique. Since the deposited SnS films form a p–n junction with the lower lying TCO, four-probe technique was used to determine the actual resistivity of the films, where the contacts were made by the spring action of the four stainless steel probes, as shown in inset of Fig. 7. A representative I–V plot for SnS film deposited on TCO glass is shown in Fig. 7.

Fig. 7 Four probe I–V plot of SnS thin film deposited on TCO glass substrate (inset: schematic of four-probe technique).

For the deposited SnS films, a linear relationship between current and voltage has been observed. Since the bottom layer of the SnS film was TCO (SnO₂:F), which is highly doped and conducting in nature, correction is essential while calculating the actual resistivity of the films, in order to nullify the interference of the TCO layer. This is done by using the manufacturer provided standard formulae and corresponding correction factor. The uncorrected resistivity (ρ°) was obtained by the equation ρ° = (V/I) × 2πs, where, V is the value of voltage at the applied current I, ‘s’ is the spacing between any two probes, which is 0.2 cm. The corrected resistivity (ρ) can be obtained by dividing ρ° with a correction factor called G₆(W/s), i.e., ρ = ρ°/[G₆(W/s)], where, W is the thickness of the deposited film. G₆ is a manufacturer defined term, where G₆(W/s) is a function, which is defined as:

The literature value for the correction factor G₆(W/s) is 0.0000019 for the films with a thickness less than 0.25 mm. Since here the thickness of the deposited SnS films are much less than 0.25 mm, this correction factor has been used in order to get the exact value of resistivity, which was found to be 1.485 × 10⁶ Ω cm. The order of resistivity of such SnS film is ideal for the use as an absorber layer of a hetero-junction solar cell.

3.6.3. Dielectric properties. Dielectric and impedance spectroscopy measurements were carried out in the frequency range 0.02 to 1000 kHz. The top contact from the SnS film was taken by applying a conducting aluminium tape with 0.4 cm² area, while the TCO glass substrate itself acted as the bottom contact. This leads to the formation of parallel plate capacitor geometry.
3.6.3.1. Variation of dielectric constant with frequency. The capacitance (C) of different films was measured with varying frequency (ω) between 0.02 kHz and 1000 kHz, under constant level of 1.0 V. The dielectric constant (ε) of the films was then calculated using the standard equation:

ε = (c × d)/(ε₀ × A)

where c is the measured capacitance, d is the thickness (in cm) of the film, A is the effective area of contact (0.4 cm²), and ε₀ is the dielectric constant of vacuum, i.e., 8.85 × 10⁻¹⁴ F cm⁻¹.²⁹ A typical exponential decay for both capacitance and dielectric constant with increasing frequency was observed, which is shown in Fig. 8a and b, respectively. This is further supported by the almost linear nature of the log–log plot of the same, as presented in Fig. 8c (for capacitance) and Fig. 8d (for dielectric constant).

Fig. 8 Variation of (a) capacitance with log frequency (b) dielectric constant with log frequency (c) log–log plot of frequency vs. capacitance (d) log–log plot of frequency vs. dielectric constant.

The significantly smaller crystallite size (∼30 nm) in the deposited films helps in increasing the number of dipoles/nano-dipoles present per unit volume in the film matrix. As the value of dielectric constant (ε) is directly dependent on the number of dipoles present per unit volume of a substance, and their orientations in the matrix, it is expected that the deposited film will show a higher value of dielectric constant, which is clearly reflected in the frequency vs. dielectric constant plot (Fig. 8b). The dielectric constant variation trend was 23.37 > 17.23 > 11.37 > 7.77 > 5.69 > 4.41 > 3.54 > 2.43 > 1.89 in the frequency range 0.02–1 kHz. At the steady condition (frequency region of 25–1000 kHz), the value of dielectric constant was found to vary within 0.277 to 0.04.

Avalanche fall in ε was observed within 25 kHz, indicating maximum distortion of the oriented dipoles in this range. Such relaxation effect of ε of SnS thin film is quite similar to other semiconducting thin films.³⁰ Beyond 25 kHz, equilibrium in distortion of dipoles with frequency was established, which is reflected in the linear part of the plot (Fig. 8b). It can be seen from Fig. 8b that the dielectric constant decreases with the increase in frequency and becomes almost constant at higher frequencies. A dielectric medium is considered to be composed of double layers, well conducting grains which are separated by poorly conducting (or resistive) grain boundaries. Under the application of external electric field, the charge carriers can easily migrate through the grains but are accumulated at the grain boundaries. This process can produce large polarization and high dielectric constant. The higher value of dielectric constant can also be explained on the basis of interfacial/space charge polarization due to inhomogeneous dielectric structure. The inhomogeneities present in the system may be porosity and grain structure. The polarization decreases with the increase in frequency and then reaches a constant value due to the fact that beyond a certain frequency of external field the metal ion (here Sn²⁺) hopping process takes place.

3.6.3.2. Dielectric loss. Fig. 9a is showing the typical nature of the variation of dielectric loss (tan δ) with increasing frequency within the range 0.02 to 1000 kHz. An avalanche breakdown in tan δ was observed within a very short range of frequency (0.02–60 kHz), where, the dielectric loss was found to fall from 3.5 to 1.29. The initial fall in the values of tan δ with increasing frequency is not unusual, since the polarization; which involves factors like electronic displacement, ionic displacement, dipole orientation and space charge displacement, is also influenced by the applied frequency field. The value of tan δ was very steady over a long range of frequency (60–1000 kHz) with the limiting value in the range 1.29 to 1.21. The positive and high value of dielectric loss implies that the SnS thin film system, which is being studied here, will not be able to store energy and will be instantaneously polarized. From the log of frequency vs. tan δ plot (Fig. 9b), the maximum dielectric loss was also found to be about 3.5 and with increasing frequency, the dielectric loss lowers down to a value of about 1.21. The relationship between dielectric loss and frequency may be explained by the Goswami–Goswami model³¹ (ESI-4†).

Fig. 9 Plot of (a) frequency vs. dielectric loss and (b) log (frequency) vs. dielectric loss for the SnS thin film.

3.6.4. Impedance measurement. Complex impedance behavior can be described by the model of a series of triple parallel R–C circuit elements that corresponds to the dielectric behavior of the grain-bulk, the grain-boundary, and the electrode-film, respectively. These are characterized by the parameters, (R_g, C_g), (R_gb, C_gb), and (R_e, C_e), where R resembles the resistance, and C resembles the capacitance of these regions. The grain-bulk impedance arises from the lattice bulk of the thin film, while the grain-boundary impedance originates from the trapped charges, due to impurities or defects in the sample, as described by Gerhardt et al.³² The grain-boundary is suggested to be a continuous phase, or a blocking layer of high resistivity, surrounding the grain-bulk region.³³ The electrode-film impedance originates from the migration/diffusion of the ionic species or charges present inside the lattice, towards the electrode-film contact and their successive accumulation.³⁴ The schematic of a complex impedance diagram is shown in Fig. 10a, in which Z′′ is plotted against Z′, where Z′ and Z′′ are the real and imaginary parts of the complex impedance Z*.

Fig. 10 (a) Model of the complex impedance spectra showing arcs due to three regions and arrow direction shows increasing frequency (b) the equivalent circuit corresponding to the three regions x, y and z.

A more satisfactory approach may be taken by considering a simple equivalent circuit as shown in Fig. 10b, which arises from the three arcs as shown in Fig. 10a. A resistance R_b was added in series to the parallel capacitance–resistance combinations, to represent the contribution of resistance from the grain-bulk interior. The grain-boundary normally exhibits higher resistance than the grain-bulk, and lower resistance than the electrode-film, i.e. R_g < R_gb < R_e (ESI-5†).

Using the equation D = tan δ = Z′′/Z′ = ε′′/ε′, we can conclude that ε′ is real part of dielectric constant that describes the stored energy, while ε′′ is the imaginary part of dielectric constant, which describes the dissipated energy. Fig. 11a is showing the frequency dependence of the real part of dielectric constant (ε′) and Fig. 11b is showing the variation of the imaginary part of dielectric constant (ε′′) with frequency. ε′ was found to decrease drastically in the low frequency zone, might be due to the polarization factors as mentioned earlier. However, a steady and limiting value of ε′ was achieved in the higher frequency zone. For the imaginary part of the dielectric constant (ε′′), the curve was found to reach a maximum with increase in frequency, indicating that the material is suitable for high frequency device application.

Fig. 11 (a) Variation of ε′ with log frequency (b) variation of ε′′ with log frequency (c) the Cole–Cole plot for the SnS thin film system showing two semicircular arcs (d) variation of ac conductivity with frequency.

The Cole–Cole plot (ε′ vs. ε′′) for the deposited SnS film is shown in Fig. 11c, in which, the variation of the real part of dielectric constant (ε′) with the imaginary part (ε′′) represents a semi circular curve over a wide range of measured frequencies. This semicircle merges and terminates on the real impedance axis at higher frequency side, indicating the presence of only bulk resistance for the sample, where the grain boundary resistance was negligibly small as no second semi-circle was observed. The absence of the series resistance in the equivalent circuit model for the sample further supports the observed result.³⁵ The semicircle starts on the real part at the lowest frequency as was also observed by Kumar et al.³⁶ The behavior of the plot is the characteristic feature of conducting nature of the sample and the absence of series capacitance in the equivalent circuit representation is envisaged. The bulk resistance of the sample can be evaluated from the low frequency intercept of the semi-circle at a particular temperature from this plot.³⁵ In general; different types of dipoles contained in the material are characterized by their own relaxation time. Therefore, the Cole–Cole plot, in general, is not exactly semicircular and varies with different degrees of distortions. The centers of this semicircle are depressed below the real axis by an angle. In an ideal case with Debye behavior, a perfect semicircle with its center lying on the real part indicates the single relaxation process. In polycrystalline materials, in addition to the arc for dielectric relaxation within the grains (bulk relaxation), another arc due to the partial or complete blocking of charge carriers at the grain boundary may also be found. Generally, the electrode processes relax at low frequencies, grain boundaries relax at intermediate frequencies and the relaxation due to the grains occurs at higher frequencies.³⁶ The observed behavior clearly indicates that the present SnS thin film has semiconductor-like properties. The increase in electrical conductivity with increasing frequency might be related to the increase in the drift mobility of electrons and holes by the hopping conduction mechanism.³⁷

3.6.5. AC conductivity. Fig. 11d shows the variation in electrical conductivity with frequency at room temperature. The ac conductivity (σ_ac) increases with increasing frequency in a typical manner, which can be determined using the following relation:

σ_ac = ε′ε₀ω tan δ

where, ω is the angular frequency and the other symbols bear their usual meaning.³⁸

The total conductivity of the system may also be expressed as:

σ_ac = σ₀(T) + σ(ω, T)

Here, first term on right hand side is the dc conductivity which is independent of frequency, and the second term is pure ac conductivity, which arises due to the electron hopping between the metal ions in the semiconductor matrix. The gradual increase in the ac conductivity with increasing frequency might be due to the enhancement of the electron hoping frequency. On the other hand, the defects in the thin film facilitate the formation of grain boundary barriers, leading to the blockage of flow of charge carriers. This in turn decreases the conductivity of the system. Among the two opposing processes, the increase in electron hoping frequency is the predominating one.

3.7. Photochemical properties

Photocatalytic behavior was studied for the SnS film deposited on TCO glass substrate using an aqueous solution of 50 ppm Xylenol Orange (XO) dye by observing the gradual decrease in intensity of its characteristic absorption maxima in the UV-vis spectrum, which occurred due to the gradual degradation of the dye in presence of SnS and light. One SnS film, with 3 cm × 3 cm active area, was placed at the bottom of a 100 ml beaker containing 50 ml of 50 ppm XO dye solution. The irradiation was carried out using a 200 W tungsten lamp (Philips) as a source of light, which was placed vertically on the reaction vessel at a distance of 1.0 cm from the top and 8 cm from the bottom. At certain time intervals, specific amount of the solution was withdrawn and the change in concentration of the XO dye was observed using the spectrophotometer.

The photocatalytic degradation was evaluated by measuring the changes in absorbance peak maxima of the characteristics peak ∼575 nm for XO at different time intervals. It was observed that there is a significant decrease in the absorption intensity with an increase in the irradiation time (Fig. 12) indicating a decrease in the concentration of the dye with time in presence of the catalyst and light. Appearance of no new absorption peak and disappearance of all existing peaks during the whole process indicates the complete photolysis in presence of the catalyst and light. No degradation of this dye was observed in dark and very slow degradation was observed in presence of light but absence of SnS films (Fig. 13). The degradation efficiency was found to be about 75% for XO dye in 250 min. The degradation rate was found to be about 0.00671 min⁻¹, which indicates pseudo-first order kinetics. To the best of our knowledge, there are no references of degradation of the dye Xylenol Orange by SnS. However, this rate is quite comparable with the degradation of other dyes like rhodamine B,³⁹ methyl orange⁴⁰ and methylene blue⁴¹ by pure SnS nanocrystals and hybrid systems. By absorbing light in the visible region, SnS produces electron–hole pairs.¹⁷ The photogenerated electrons (e⁻) occupy the conduction band of SnS and are scavenged by O₂ to produce O₂˙⁻ anion radicals which on protonation yield HOO˙ radicals. On the other hand, the holes (h⁺) in the valence band reacts with H₂O (or –OH ions present in H₂O) to produce highly reactive species ˙OH radical. These ˙OH radicals and O₂˙⁻ anion radicals thus produced are responsible for the complete mineralization of the dye by reductive and oxidative degradation routes respectively, leading to the products like CO₂ and H₂O. These reactions can be summarized as follows:^42,43

SnS + hv → e⁻_CB + h⁺_VB

e⁻_CB + O₂ → O₂˙⁻

O₂˙⁻ + H⁺ → ˙OOH

˙OOH → H₂O₂ + O₂

H₂O₂ + e⁻ → ˙OH + ⁻OH

h⁺_VB + ⁻OH/H₂O → ˙OH

˙OH + dye → Degraded products (reductive degradation)

O₂˙⁻ + dye → Degraded products (oxidative degradation)

Fig. 12 UV-vis spectra showing the quenching of intensity of the characteristics peaks of the XO dye solution in presence of the SnS film under illumination.

Fig. 13 Plot of irradiation time vs. ln(C_t/C₀) for the determination of dye degradation rate and kinetics.

In order to determine the stability and repeatability of the proposed SnS thin film based catalyst, the same film was used for 5 consecutive times to degrade XO solution and the result is summarized in Fig. 14. Very small change in the degradation efficiency (75% for 1^st cycle and 70% for 5^th cycle) was observed from the figure even after 5 successive cycles of XO degradation, indicating robust nature of the SnS thin films deposited by this technique.

Fig. 14 Plot of photodecomposition cycle vs. degradation efficiency of the SnS film towards XO solution.

3.8. Photoelectrochemical properties

One of the most important aspects of utilizing solar energy is its conversion from solar radiation to electrical energy, which can be done using photoelectrochemical (PEC) cells most easily. Such PEC cells have many advantages over the solid state photovoltaics like they are not sensitive to the defects in semiconductor and the solid/liquid junction can be easily fabricated leading to cost minimization. In a PEC cell, both the semiconductor electrode and the counter electrode are immerged in the redox electrolyte. The incident light excites the semiconductor electrode and the photo generated electrons and holes are separated in the space charge region.^44,45

For the study of heterojunction solid/liquid solar cell, i.e. PEC cells, a two-electrode cell configuration was used, in which the counter electrode was Pt and the working electrode was p-SnS thin film. These two electrodes were immersed in a redox electrolyte containing 0.1 (M) KCl, 0.1 (M) K₄[Fe(CN)₆] and 0.1 (M) K₃[Fe(CN)₆] in a beaker and were connected externally to the KEITHLEY-4200 semiconductor characterization system for measuring the PEC performances. Prior to measurement, the solution was purged with pure nitrogen and an atmosphere of nitrogen was maintained during the experiment. The distance between the semiconductor photoelectrode and counter electrode was kept constant at 0.5 cm. The active area of the SnS film and the light intensity on the semiconductor surface were 1.0 cm² and 100 mW cm⁻², respectively. In order to avoid disturbances from stray light, all photoelectrochemical experiments were carried out in a light blocking cabinet.

In this redox couple, the SnS photo-anode based PEC cell showed no efficiency under dark (Fig. 15a), but significant short-circuit photocurrent (I_sc) of 0.0108 A with an open circuit voltage (V_oc) of 0.438 V was observed when the film was illuminated (Fig. 15b). The fill factor (FF) was calculated to be about 44% leading to an overall cell efficiency (η) of 2.08%. The value of efficiency we have obtained is notably higher than many previous reports of SnS based PEC cells which are summarized in Table 2. The enhanced photoelectrochemical efficiency of such galvanically deposited SnS films might be attributed to the high dielectric stability and suitable band gap energy to absorb photons, as discussed in the previous sections. Fig. 15c is representing the maximum power (P_max = I_max × V_max) output of the fabricated cell with respect to voltage, which was found to be about −0.00207 W. The output of this PEC cell was almost constant for more than 1000 hours, which means, such galvanically deposited SnS films are stable chemically as well as photochemically and they may be considered as a suitable material for PEC cell fabrication.

Fig. 15 (a) Dark I–V and (b) light I–V curve of the PEC cell fabricated with galvanically deposited p-SnS thin film as photo-anode in presence of Fe(II)/Fe(III) redox couple.

Table 2 Comparison of performance of different SnS based photoelectrochemical cells

Heterojunction	Voc (mV)	Isc (mA cm^−2)	FF	η (%)	References
TCO/SnS	438	0.0108	44	2.08	This work
SnS/TiO2	471	0.30	0.71	0.1	46
ZnO/SnS	120	0.04	0.33	0.003	47
SnO2:F/CdS/SnS/Cu2SnS3	340	6.00	Not available	Not available	48
Cd0.87Zn0.13S/SnS	288	9.16	0.27	0.71	49
CdS/SnS	260	9.6	0.53	1.30	16
SnS2/SnS	350	1.5	Not available	Not available	50
CdO/SnS	200	0.054	Not available	Not available	51
Cd2SnO4/SnS	230	0.039	Not available	Not available	51
SnO2:F/SnS	152	0.123	Not available	Not available	51

---

4. Conclusion

Polycrystalline SnS thin films with orthorhombic structure were deposited on TCO coated glass substrate by a modified electrochemical technique. The morphology of the deposited films was moderately dense with a regular distribution of the spherical particles. The material showed direct band gap energy at around 1.5 eV, typical for SnS. I–V measurements showed diode characteristic of the SnS/TCO junction. The order of resistivity (10⁶ Ω cm) for the SnS film was found to be ideal to be used as an absorber layer of a hetero-junction solar cell. The material was found to show a steady dielectric constant in the higher frequency region, making it suitable for high frequency device applications. The presence of semicircular arcs in the Cole–Cole plot which merges and terminates on the real impedance axis at higher frequency side, indicates the presence of bulk only resistance. A shorter relaxation time of polarization also indicates that the imaginary dielectric constant reaches the maximum at higher frequency. The ac conductivity of the material was also found to be influenced by the electron hoping process, which predominates over the blockage of formation of charge carriers due to the defects in the semiconductor matrix and subsequent grain boundary formation. The band-gap energy, resistivity, dielectric constant and relaxation time make this material and ideal absorber layer for efficient photochemical and photoelectrochemical behavior.

Acknowledgements

The authors would like to acknowledge Mr Subhais Das & Mr Sudhir Ghosh of JU for their technical support. Authors are thankful to MHRD-India and UGC-SAP (India) for providing instrumental facilities to the Department of Chemistry, IIEST, Shibpur.

Notes and references

1. E. C. Greyson, J. E. Barton and T. W. Odom, Small, 2006, 2, 368–371.
2. J. B. Johnson, H. Jones, B. S. Latham, J. D. Parker, R. D. Engelken and C. Barber, Semicond. Sci. Technol., 1999, 14, 501–507.
3. S. Cheng, Y. Chen, C. Huang and G. Chen, Thin Solid Films, 2006, 500, 96–100.
4. M. Devika, N. Reedy, K. Gunasekhar, E. Gopal and K. Reddy, Semicond. Sci. Technol., 2006, 21, 1125–1131.
5. N. Sato, M. Ichimura, E. Araia and Y. Yamazaki, Sol. Energy Mater. Sol. Cells, 2005, 85, 153–165.
6. O. E. Ogah, G. Zoppi, I. Forbes and R. W. Miles, Thin Solid Films, 2009, 517, 2485–2488.
7. H. Noguchi, A. Setiyadi, H. Tanamura, T. Nagatomo and O. Omoto, Sol. Energy Mater. Sol. Cells, 1994, 35, 325–331.
8. E. Guneri, C. Gumus, F. Mansur and F. Kirmizigul, J. Optoelectron. Adv. Mater., 2009, 3, 383–389.
9. M. T. S. Nair and P. K. Nair, Semicond. Sci. Technol., 1991, 6, 132–134.
10. M. Ichimura, K. Takeuchi, Y. Ono and E. Arai, Thin Solid Films, 2000, 361, 98–101.
11. W. G. Pu, Z. Z. Lin, Z. W. Ming, G. X. Hong and C. W. Qun, Proceeding of the 1st world conference on photovoltaic energy conversion, Hawaii, December, 1994, pp. 365–638.
12. B. Ghosh, M. Das, P. Banerjee and S. Das, Appl. Surf. Sci., 2008, 254, 6436–6440.
13. G. H. Yue, W. Wang, L. S. Wang, X. Wang, P. X. Yan, Y. Chen and D. L. Peng, J. Alloy Compd., 2009, 474, 445–449.
14. A. Tanusevski and D. Poelman, Sol. Energy Mater. Sol. Cells, 2003, 80, 297–303.
15. M. Ristov, G. Sinadinovski, I. Gro and M. Mitreski, Thin Solid Films, 1989, 173, 53–58.
16. K. T. R. Reddy, N. K. Reddy and R. W. Miles, Sol. Energy Mater. Sol. Cells, 2006, 90, 3041–3046.
17. A. Umar, M. S. Akhtar, G. N. Dar, M. Abaker, A. Al-Hajri and S. Baskoutas, Talanta, 2013, 114, 183–190.
18. B. Subramanian, C. Sanjeeviraja and M. Jayachandran, Mater. Chem. Phys., 2001, 71, 40–46.
19. S. G. Patil and R. H. Tredgold, J. Phys. D: Appl. Phys., 1971, 4, 718–720.
20. P. K. Nair and M. T. S. Nair, J. Phys. D: Appl. Phys., 1991, 24, 83–86.
21. A. Mondal, N. Mukherjee and S. K. Bhar, Mater. Lett., 2006, 60, 1748–1752.
22. B. Chakraborty, B. B. Show, S. Jana, B. C. Mitra, S. K. Maji, B. Adhikary, N. Mukherjee and A. Mondal, Electrochim. Acta, 2013, 94, 7–15.
23. J. R. S. Brownson, C. Georges, G. Larramona, A. Jacob, B. Delatouche and C. L. Clement, J. Electrochem. Soc., 2008, 155(1), D40–D46.
24. G. Mattock, J. Chem. Soc., 1954, 989–997.
25. B. D. Cullity, Elements of X-ray diffraction, Addision-Wesley, Massachusetts, 1st edn, 1956.
26. G. Willeke, R. Dasbach, B. Sailer and E. Bucher, Thin Solid Films, 1992, 213, 271–276.
27. G. B. Williamson and R. C. Smallmam, Philos. Mag., 1956, 1, 34–46.
28. Y. T. Lin, J. B. Shi, Y. C. Chen, C. J. Chen and P. F. Wu, Nanoscale Res. Lett., 2009, 4, 694–698.
29. N. Mukherjee and A. Mondal, J. Electron. Mater., 2010, 39, 1177–1185.
30. D. Ghosh, P. S. Sardar, M. Biswas, A. Mondal and N. Mukherjee, Mater. Chem. Phys., 2010, 123, 9–12.
31. K. Deraman, Electrical and optical properties of tin sulphide thin films, Ph.D thesis, Univ. Teknol, Malaysia, 1998.
32. R. Gerhardt and A. S. Nowick, J. Am. Ceram. Soc., 1986, 69, 641–646.
33. D. Y. Wang and A. S. Nowick, J. Solid State Chem., 1980, 35, 325–333.
34. O. K. Varghese and L. K. Malhortra, J. Appl. Phys., 2000, 87, 7457–7466.
35. G. Rambabu, N. Anantharamulu, K. Koteshwara Rao, G. Prasad and M. Vithal, Mater. Res. Bull., 2008, 43, 1509–1518.
36. V. Prashanth Kumar, Y. S. Reddy, P. Kistaiah, G. Prasad and C. Vishnuvardhan Reddy, Mater. Chem. Phys., 2008, 112, 711–718.
37. J. E. Bao, J. Zhou, Z. X. Yue, L. T. Li and Z. L. Gui, Mater. Sci. Eng., B, 2003, 99, 98–101.
38. A. Azam, A. S. Ahmed, M. Chaman and A. H. Naqvi, J. Appl. Phys., 2010, 108, 94329–94336.
39. P. Tang, H. Chen, F. Cao, G. Pan, K. Wang, M. Xu and Y. Tong, Mater. Lett., 2011, 65, 450–452.
40. C. Yang, W. Wang, Z. Shan and F. Huang, J. Solid State Chem., 2009, 182, 807–812.
41. L. Zhu, Z.-D. Meng and W.-C. Oh, J. Korean Ceram. Soc., 2012, 49, 155–160.
42. K. Vinodgopal, I. Bedja and P. V. Kamat, Chem. Mater., 1996, 8, 2180–2187.
43. W. Zhao, Y. Wang, Y. Yang, J. Tang and Y. Yang, Appl. Catal. B, 2012, 115–116, 90–99.
44. D. Wei and G. Amaratunga, Int. J. Electrochem. Sci., 2007, 2, 897–912.
45. N. Chandra, J. K. Leland and A. J. Bard, J. Electrochem. Soc., 1987, 134(1), 76–80.
46. Y. Wang, H. Gong, B. Fan and G. Hu, J. Phys. Chem. C, 2010, 114(7), 3256–3259.
47. B. Ghosh, M. Das, P. Banerjee and S. Das, Semicond. Sci. Technol., 2009, 24, 025024/1–025024/7.
48. D. Avellaneda, M. T. S. Nair and P. K. Nair, Thin Solid Films, 2009, 517, 2500–2502.
49. M. Gunasekaran and M. Ichimura, Sol. Energy Mater. Sol. Cells, 2007, 91, 774–778.
50. A. Sanchez-Juarez, A. Tiburcio-Silver and A. Ortiz, Thin Solid Films, 2005, 480–481, 452–456.
51. M. Ristov, G. Sinadinovski, M. Mitreski and M. Ristova, Sol. Energy Mater. Sol. Cells, 2001, 69, 17–24.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra11140k

---