Synthesis

The mixed solution for the synthesis of tin sulfide NPs was prepared by adding 40 mL of 0.1 m CH₃COONH₄ to 40 mL of 0.1 m SnCl₂. Then 20 mL of 0.3 m Na₂S₂O₃ was added to the solution. Finally, the solution pH was adjusted to five with the addition of drops of ammonia. The mixed solution was stirred for 1 min. All these processes were carried out at room temperature without any specific conditions. In order to make the electrolyte for doping, an aqueous solution of InCl₂ at a concentration of ~1 × 10⁻³ m was prepared in a separate beaker. By adding different quantities of this solution (1, 3, and 5 cm³) to different beakers, separate experiments were carried out. The samples were called In(0), In(1), In(2), and In(3) for the undoped and doped samples with different amounts of the solution, respectively. Ultrasonic waves were introduced at frequencies of 20 kHz using titanium probe (diameter ~12 mm) (Top Sonics UPH-400, 20 kHz–400 W, Ultrasonic Technology Development Co., Iran). For the synthesis, the probe was kept 1 cm inside the solution, and the sonicator was turned on for 6 s and off for 3 s, with this procedure repeated 100 times for all samples. After the deposition, each solution was removed from the beaker, using the filtering paper gently, and several times washed by distilled water, and dried in air.

Characterization

The NPs were crystalline characterized using an x-ray diffractometer (XRD, PAN Analytica) with a monochromatized Cu Kα radiation source (1.5406 Å). A field-emission scanning electron microscope (FESEM, TeScan, Mira 3-XMU) was used for morphological studies. The elemental composition was obtained with an energy-dispersive x-ray spectrometer (EDS) attached to the same FESEM. The optical properties were investigated by means of the UV–vis analysis at room temperature. The UV–vis spectra were recorded by a UV–vis spectrometer (Hatch DR 5000 UV–Vis Spectrometer) over a wavelength range of 300–1000 nm.

Electrical and solar cell experiment

The tin sulfide NPs were deposited in the form of film on a fluorine doped tin oxide (FTO) glass substrate using the chitosan binder. For this purpose, 1 × 2 cm² FTO substrates were ultrasonically cleaned by acetone, ethanol, and distilled water for 10 min in each solution at ultrasonic bath and finally dried in air. The 0.6 mL chitosan binder with 10 mg NPs mixed and on the FTO substrate at a 0.5 × 0.5 cm² window that created by parafilm and dried in air for 24 h. Ag paste was used as a cathode electrode according to Fig. 1 for solar characterization. The effective surface of the samples exposed to the light is 0.25 cm². The Mott–Schottky characteristics was obtained according to our previous report.^[ Reference Jamali-Sheini, Cheraghizade and Yousefi ^{20 ]} The solar cell measurement and external quantum efficiency (EQE) were carried out under 100 mW/cm² (1.5 Air Mass) illumination from a solar simulator (Solar cell simulator IIIS-200+, Nanosat Co., Iran). A 100 W xenon lamp served as a light source, and the intensity of the light was calibrated using a standard silicon solar cell. Photocurrent measurements were carried out by a potentiostat (Autolab, A3ut71167, Netherlands) for plot of J–V and EQE.

Figure 1.

Schematic representation of fabricated solar cells.

Crystalline and morphological properties

XRD patterns of the tin sulfide NPs are shown in Fig. 2. All the samples were crystalline and of the orthorhombic phase of SnS and Sn₂S₃ (JCPDS card Nos 14-0620 and 14-0619) as well as the hexagonal phase of SnS₂ (JCPDS card No. 83-170).^{[ 21 ]} The existence of broad peaks in all XRD diffractogram of samples indicated the formation of nano-sized particles. The volume percent of different phases present in the samples was determined from the integrated intensities of the diffraction peaks and using the following expressions reported in the literature.^[ Reference Wang, Zhao, Xu, Fu and Wang ^{22 ,} Reference Basahel, Ali, Narasimharao, Bagabas and Mokhtar ^{23 ]} The result was obtained for all samples and summarized in Table I. Therefore, it can be observed that the main phase is SnS in the crystal structure.

Figure 2.

XRD patterns of samples.

Table I.

The relative percent of different phases at tin sulfide NPs.

The FESEM images of the tin sulfide NPs are demonstrated in Fig. 3. The NPs show a flower-like morphology. To estimate NPs’ size, a particle size histogram has been calculated and plotted. Figure 4 reveals the particle size histogram of tin sulfide NPs. These graphs represent a particle size <100 nm for all samples. First, by adding In(1), the particle size decreases, and then by increasing indium percentage to In(2) and In(3), the particle size increases, too. An elemental purity of the tin sulfide NPs was tested by the EDS technique, and the typical spectrum of all samples is presented in Fig. 5. It is evident that In percentage rises according to our expectation.

Figure 3.

FESEM images of (a) In(0), (b) In(1), (c) In(2), and (d) In(3) samples.

Figure 4.

Particle size histogram of (a) In(0), (b) In(1), (c) In(2), and (d) In(3) samples.

Figure 5.

EDS spectra of (a) In(0), (b) In(1), (c) In(2), and (d) In(3) samples.

Sonochemical mechanism growth of tin sulfide NPs begin by providing Sn²⁺ and S²⁻ cations and anions that dissociation in SnCl₂ and Na₂S₂O₃, respectively. CH₃COONH₄ can be hydrolyzed in aqueous solution to produce NH₃ and generates hydroxide anions. Finally, the formation of SnS nuclei could occur. In the presence of sonication energy, these processes can be accelerated and have a tendency to produce flower-like particles. In the presence of InCl₂ as dopant source, In²⁺ cations also dissociate in solution and incorporated with sulfur anions.

Optical properties

The UV–vis absorbance spectra of the tin sulfide NPs appear in Fig. 6. Tin sulfide NPs have high absorbance in the range of UV, and the absorbance is about 74%, 88%, 84%, and 78% for In(0), In(1), In(2), and In(3) SnS NPs, respectively. The absorbance raise rapidly, using indium dopants. In(1) NPs have relatively stronger absorbance intensity in the UV and visible regions compared with the other samples. This could be due to the fact that these NPs have the smallest size in comparison with the other samples.

Figure 6.

UV–vis spectra of the samples.

The optical band gaps of the tin sulfide NPs could be evaluated via Kubelka–Munk equation.^[ Reference Jamali-Sheini, Yousefi, Ali Bakr, Cheraghizade, Sookhakian and Huang ^{24 ]} Figure 7 shows the curves of (αhν)² versus hν (Tauc plots) of the tin sulfide NPs. The curves have a good straight-line fit with higher-energy range above the absorbance edge, indicating a direct optical transition near the absorbance edge. In addition, form the Tauc plots and considering the energy band gap values of SnS (1–2 eV) and SnS₂ (1.9–2.4 eV), we believe that the dominate phase could be SnS.

Figure 7.

Tauc plot of (a) In(0), (b) In(1), (c) In(2), and (d) In(3) samples.

Based on Fig. 7, the direct E _g of the tin sulfide NPs have been first a blue shift and then show a red shift, for In(2) and In(3) samples. According to the literature survey in doped semiconductors, optical energy band gaps of In-doped tin sulfide NPs can be achieved by the following relation^[ Reference Dezfuly, Yousefi and Jamali-Sheini ^{25 ,} Reference Jagadish ^{26 ]}:

(1) $$E_{\rm g}\;({\rm S}{\rm n}_{1 - x}{\rm I}{\rm n}_x{\rm S}) = (1 - x) \times E_{\rm g}({\rm SnS}) + x \times E_{\rm g}{\rm I}{\rm n}_x{\rm S}{\rm.} $$

Thus, knowing that the optical band gap energy of In-SnS NPs is bigger than that of SnS and considering that the particle size is smaller, a blue shift for In(1) SnS NPs is reasonable. However, the red shift in optical energy band gaps of In(2) and In(3) samples is probably due to the merging of indium level with host bands or the band shrinkage by the presence of dopant ions. This increase with indium dopant has also been observed and reported in other works.^[ Reference Chaki, Chaudhary and Deshpande ^{17 ,} Reference Reghima, Akkari, Guasch and Kamoun-Turki ^{19 ]}

Electrical and photovoltaic performances

The Mott–Schottky method, is useful way to check the electrical characteristic of the samples such as doping concentration, flat-band potential, etc by measuring the capacitance of the cell containing the semiconductor as a function of applied voltage.^[ Reference Jamali-Sheini, Cheraghizade and Yousefi ^{20 ]} In addition, the Mott–Schottky equation has an essential role in estimating the type of conductivity (i.e., n- or p-type). Using the equations from our previous publication,^[ Reference Jamali-Sheini, Cheraghizade and Yousefi ^{20 ]} the calculations show all samples exhibit p-type conductivity with different values of majority charge carrier (Table II). It is important to note that In-doped NPs have higher charge carrier concentration, showing that indium as cationic dopants for SnS NPs.

Table II.

Hole concentration of samples.

Figure 8 shows the J–V curves of solar cells fabricated from the tin sulfide NPs. The results (Table III) show that the solar cell with In(1) has higher efficiency and power efficiency conversion. In-doped tin sulfide NP-based solar cells have a relatively high open-circuit voltage. In the present case, doping level and size of particles have key parameters for obtaining the high open-circuit voltage. For comparing our result with other works, that use a thin film for fabrication of SnS-based solar cells, we collected a number of other works that have been shown in Table IV. The enhanced results were obtained for In-doped tin sulfide NPs among them. In addition, other solar cell parameters such as the series and shunt resistance (R_ss & R_sh) were calculated for solar cell, which was fabricated from the tin sulfide NPs^[ Reference Jamali-Sheini, Cheraghizade and Yousefi ^{20 ]} and are shown in Table III. It can be observed that the series resistance increases, whereas the shunt resistance first increases by using a minimum concentration of indium dopants and then decreases by using higher concentration of indium dopant. The series resistance is closely related to the intrinsic resistance, morphology, and size of the semiconductor NPs. Furthermore, the shunt resistance is correlated with the amount and the character of the impurities and defects in the active semiconductor layer because impurities and defects cause charge recombination and leakage current.^[ Reference Kim, Kim and Kim ^{27 ]} These improvements in photovoltaic performance of In-doped SnS-based solar cells are due to the increases in the charge carrier and effective reductions in the charge recombination.^[ Reference Wang, Qu, Zeng, Liu, Zhang, Tan and Jin ^{28 ]}

Figure 8.

J–V curves of the tin sulfide NP-based solar cells.

Table III.

Photovoltaic parameters of tin sulfide NPs as solar cells.

Table IV.

Reported experimental efficiencies of single-layer SnS-based solar cells.

Figure 9 shows the EQE as a function of excitation wavelength for un- and In-doped SnS NP-based solar cells. A maximum and minimum of 32.5% and 20% EQE achieved for In(1) and In(0) samples, respectively. Improving in EQE of In-doped samples is due to reducing recombination by decreasing the number of defects such as tin vacancy.^[ Reference Steinmann, Jaramillo, Hartman, Chakraborty, Brandt, Poindexter, Lee, Sun, Polizzotti, Park, Gordon and Buonassisi ^{11 ]} Generally, other parameters that can improve the efficiency and EQE of SnS-based solar cells are including band alignment, absence of secondary phases, short minority carrier lifetime and diffusion length, solar cell configuration, recombination at the junction, doping level, back contact barrier and front contact/buffer, thickness of film, size of particles/structures, band gap, electron affinity, crystal structure, electrical nature, doping, and deposition technique.^[ Reference Reddy, Gedi, Park, Miles and Kt ^{29 ,} Reference Andrade-Arvizu, Courel-Piedrahita and Vigil-Galán ^{30 ]}

Figure 9.

EQE spectra of the SnS NP-based solar cells.