Structural, Optical, Photoelectrochemical, and Electronic Properties of the Photocathode CuS and the Efficient CuS/CdS Heterojunction.

We report a facile synthesis of CuS and CdS nanoparticles using a cheap solution-processed chemical bath method forming heterogeneous nucleation. The optical, structural, photoelectrochemical (PEC), and electronic properties are studied by implementing relevant experimental techniques. The estimated optical band gap of ∼2.10 eV of CuS designates potential application in inexpensive photocatalysis and solar cells. Further, the valence band and conduction band positions of CuS and CdS are evaluated using cyclic voltammetry curves. Narrow conduction and valence band offset potentials measured at the CuS/CdS heterojunction are encouraging factors for the PEC application. The electronic properties are supported by the current density vs potential plots (J-V) with an improved short-circuit current density of 0.71 mA cm-2 for the heterojunction compared to bare CuS showing 0.15 μA cm-2. The determined PCE of the heterojunction CuS/CdS is 0.65%.

Introduction

Environmental concerns and the increasing energy crisis caused by the extensive use of fossil fuels promote the development in the field of energy conversion and storage.[1,2] In the energy conversion field, a large number of photovoltaic (PV) materials gained attention and grew rapidly as a potential source of energy generation. The increasing demand for efficient and cost-effective solar cells promoted an immense amount of search for stable, nontoxic, and earth-abundant solar absorber materials. Chalcogenide materials such as CdTe and CIGS (Cu2InGaSe4) have been investigated with photoconversion efficiencies (PCE) of 22.1 and 22.6%, respectively.[3,4] However, the elements Ga, In, and Cd are associated with toxicity, and their higher cost limits their future as potential large-scale PV materials.[3,5] In contrast, quaternary chalcopyrites/chalcogenides Cu2MSnX4 (M = Fe, Co, Ni, or Fe and X = S and Se) are under intensive research as potential nontoxic and earth-abundant alternatives and show superior optoelectronic properties, suitable direct band gaps, and efficient charge mobility resulting in high photoconversion efficiency.[6−24]

Binary CuS is an alternative to rare-earth element-based semiconductors with an inexpensive and bulk synthesis method. CuS is a p-type semiconductor having a direct and suitable band gap (1.6–2.2 eV) for optoelectronic and PEC applications. Yuan et al. studied dye-sensitized CuS solar cells and reported a short-circuit current density (Jsc) of 2 μA cm–2 and an open-circuit voltage (Voc) of 0.17 V under a xenon lamp (100 mW cm–2).[25] CuSbS2/CdS thin films formed by annealing CuS/Sb2S3 solar cells reported by Medina-Montes et al. showed a Jsc of 0.55 mA and a Voc of 0.07 V under an illumination intensity of 100 mW cm–2.[26] A high efficiency of 5.03% was reported by Patil et al. for a solution-processed CuS counter electrode-based dye-sensitized solar cell.[27] A lightweight p-type semiconductor CuS shows huge potential and promotes more research in the field of photovoltaics.

CuS nanoparticle synthesis was reported by distinct methods such as a hydrothermal method, a solution-processed one-pot method, chemical bath deposition, pulsed laser ablation, and water-soluble CuS synthesis.[27−31] One-pot solution-processed and chemical bath methods are energy-efficient and low-cost synthesis methods.

Here, we synthesized binary chalcopyrite CuS and CdS nanocrystals of hexagonal structure using cost-effective chemical bath synthesis techniques. The structure, optical properties, and surface morphology of the as-prepared CuS and CdS nanocrystals were characterized by X-ray diffraction (XRD), ultraviolet–visible spectroscopy, Raman spectroscopy, and scanning electron microscopy techniques. Heterojunction interface analysis was carried out by electrochemical impedance spectroscopy and Mott–Schottky plots. The semiconductor heterojunction CuS and CdS interface band diagram including the conduction band offset (CBO) and the valence band offset (VBO) was evaluated using cyclic voltammetric curves and photoelectrochemical analysis using a Metrohm Autolab series potentiostat/galvanostat workstation.

Experimental Section

Synthesis of CuS by a Colloidal Method

Chemicals

Copper sulfate (CuSO4), thiourea [SC(NH2)2], ammonia solution (35%), distilled water, acetone, and toluene were purchased from Merck (Sigma Aldrich). All chemicals were directly used without extra purification.

Synthesis of CuS NPs

CuS NPs were prepared by the colloidal method. For the synthesis, CuSO4 (25 mmol) and SC(NH2)2 (0.075 mol) were dissolved in a beaker consisting of distilled water (100 mL). The temperature was raised to 60 °C, and the solutions containing precursors were mixed; then, the temperature was raised to 80 °C leading to white precipitation. Ammonia solution (2 mL, 35%) was poured into the solution leading to black precipitation, and pH 10 was observed. The precipitate was collected by centrifugation and cleaned with acetone/toluene (3:1) solution. The obtained sample was dried under an IR lamp and used for further characterization. The synthesis setup is shown in Scheme .

Scheme 1

Schematic Representation of the Chemical Bath Deposition Method of the CuS/CdS Thin Film on the FTO Substrate

Synthesis of CdS by Chemical Bath Deposition

Cadmium acetate (Cd(CH3CO2)2), thiourea [SC(NH2)2], and ammonia solution were purchased from Merck (Sigma Aldrich).

Synthesis of CdS NPs

CdS NPs were synthesized by CBD. The process consisted two different beakers containing distilled water and the precursors Cd(CH3CO2)2 (0.025 mol) and thiourea [SC(NH2)2] (0.075 mol) at a temperature of 60 °C. Once 60 °C was reached, precursor solutions were mixed, and the ammonia solution of 2 mL (35%) was poured, and pH 10 was observed. The solution turned yellow; then, FTO glass was hung inside the solution for deposition of CdS NPs.

Thin-Film Formation of CuS and CuS/CdS by Drop-Casting and CBD

FTO glass substrates were used for thin films of CuS and CuS/CdS. CuS NPs were dissolved in 1-methyl-2-pyrrolidone. The solution was drop-casted on FTO glass and dried under an IR lamp. CuS thin films were poured in CdS solution prepared as aforementioned for 5 min to form the CuS/CdS heterojunction and used for further analysis.

Results and Discussion

Structural Properties of Pure CuS and CdS (XRD and Raman Spectroscopy)

Structural properties of CuS and CdS were obtained by X-ray diffraction (XRD) pattern analysis. Figure a shows the XRD pattern of CuS NPs with the sharp diffraction peaks positioned at the following 2θ values with their corresponding planes: 27.54 (101), 29.14 (102), 31.74 (103), 32.65 (006), 47.78 (110), 52.39 (108), and 59.10° (116) showing high crystallinity. The diffraction pattern matched well with JCPDS no. 06-0464 corresponding to the hexagonal structure.[32,33] The crystalline sizes evaluated were D110 = 9.3 nm for the plane (110) and D102 = 3.1 nm, which is smaller than the report of 17.9 nm by Patil et al.[27] for the plane (102) using the Debye–Scherrer equation D = Kλ/B × cos θ, where B is the full width at half-maximum intensity (FWHM) in radians, θ is the Bragg angle, K is the shape factor, i.e., 0.89, and λ is the wavelength of the X-ray source, i.e., 1.54 Å. Moreover, with the similar variable, using the Debye–Scherrer equation, the interplanar spacings (d) evaluated by Bragg’s law, d = λ/(2 × sin θ), for planes were d110 = 0.19 nm and d102 = 0.3 nm. For the hexagonal structure, lattice parameters of CuS NPs by XRD pattern analysis are a = 0.38 nm, b = 0.38 nm, and c = 1.82 nm (a = 3.796 Å and b = c = 16.344 Å as per JCPDS no. 06-0464) evaluated by the following expression:[34]

Figure 1

(a) XRD pattern, (b) Raman spectra, (c) UV–visible absorbance plot, and (d) Tauc’s plot of CuS nanoparticles synthesized by a chemical route.

Moreover, the volume of the hexagonal[35] CuS cell V = = 227.59 Å3. Similarly, for CdS NPs, the diffraction peaks are positioned at 2θ values with corresponding planes of 25.81 (111), 27.41 (002), 29.21 (101), 37.33 (102), 44.85 (110), 49.15 (103), and 52.96° (112) matched with JCPDS no. 41-1049.[36] The diffraction pattern affirms that the structure of CdS NPs is hexagonal, as displayed in Figure S1a. Moreover, the crystalline sizes evaluated of CdS NPs for the planes representing (110) and (002) are D110 = 6.9 nm and D002 = 7.7 nm, respectively. However, the interplanar spacings for the planes (110) and (002) are d110 = 0.2 nm and d002 = 0.32 nm, respectively. The lattice parameters evaluated for CdS NPs are a = 0.4 nm, b = 0.404 nm, and c = 6.154 nm, and the volume of the hexagonal CdS cell is 852.70 Å3. Diffraction patterns of CuS and CdS NPs match the previous reports.

Both CuS and CdS have distinct structures and hence vibrational modes; therefore, further identification was obtained by Raman spectroscopy of CuS and CdS. Figure b shows Raman spectra of CuS having the evident peak of the vibrational stretching mode at 474 cm–1, which is designated for the covalent S–S bond, and the smaller peak designated for the Cu–S bond vibration at 268 cm–1. Raman spectra of CdS (Figure S1b) show significant peaks at 295 and 592 cm–1 corresponding to 1LO (optical phonon mode) and 2LO (first overtone mode); however, a smaller peak at 892 cm–1 is assigned to 3LO (second overtone mode). Raman spectra of CuS and CdS best match with previous reports with no secondary structures.[37,38]

Optical Properties of CuS and CdS

Optical absorption of materials is essential for photovoltaic application formalizations. Figure c and Figure S1c show the UV–visible absorption spectra of CuS and CdS in the wavelength ranges of 300–900 and 300–780 nm; Figure d and Figure S1d show the Tauc plots of CuS and CdS NPs, in which the interception on the x-axis gives the band gap of the sample. Tauc’s plot was obtained by the following expression:[39]in which h is Planck’s constant, ν is the frequency of radiation, Eg is the optical energy band gap, and α is the absorption coefficient. Optical band gaps evaluated by Tauc’s plot are 2.10 and 2.07 eV for CuS and CdS, respectively.

Morphological Properties of CuS and CdS NPs

Figure a,b shows the SEM images of CuS NPs showing bulk and nonuniform crystalline growth with lengths observed at 1–5 μm in size. However, the SEM images of CdS displayed in Figure c,d show the mixture of the agglomeration flowerlike morphology.

Figure 2

SEM images (a,b) corresponding to CuS and images (c,d) corresponding to CdS NPs were observed under the magnifications of 3000× and 10,000×.

Photoelectrochemical Measurements of CuS, CdS, and CuS/CdS

The electrochemical measurements such as electrochemical impedance spectroscopy (EIS), cyclic voltammetry (CV), linear sweep voltammetry (LSV), and impedance potential scan for Mott–Schottky (M–S) plots were carried out using the electrochemical workstation Metrohm potentiostat. For illumination, halogen lamps were used with an intensity of 10.1 mW cm–2. The measurements were carried out in a standard three-electrode system consisting of a reference electrode, i.e., a saturated calomel electrode [SCE = 0.244 V vs the normal hydrogen electrode (NHE)], a counter electrode (platinum plate), a working electrode (CuS, CdS, and CuS/CdS) in the electrolyte 0.5 M Na2SO4 at pH 7. The arrangement of PEC is shown in Scheme . The potentials were converted to a real hydrogen electrode (RHE) for LSV and M–S plots using the following expression:[40]

Scheme 2

Standard Three-Electrode Photoelectrochemical Cell with CuS/CdS as a Working Electrode (WE), a Saturated Calomel Electrode (SCE) as a Reference Electrode (RE), and Platinum as a Counter Electrode (CE)

EIS, M–S, and LSV Analysis

Electrochemical Impedance Spectroscopy

EIS of CuS and its heterojunction with the n-type semiconductor CdS was studied for the determination of impedance and recombination time by the charge transfer between the frequencies ranging from 0.1 Hz to 100 kHz. Figure a,b displays the Bode phase plot and the Nyquist plot of CuS and the heterojunction CuS/CdS, and Figure S2a,b displays the Bode phase and Nyquist plots for bare CdS. The determination of charge carrier concentration was evaluated using the Bode phase plot. Under the electrochemical kinetics, the PEC cell has different possible conditions of concentrations of ions due to varying potentials applied. This change in kinetics results in a change in ions transfer & electron transfer in the electrons at the electrode/electrolyte interface, which can be observed through impedance analysis. The Nyquist plot and the Bode phase plot were analyzed using EIS data. The Bode phase plot resembles the change in phase angle as compared to the input ac sine waves. The charge carrier lifetime (τ) is evaluated by the Bode phase using the expression given below:[41]

Figure 3

(a) Bode phase plot, (b) Nyquist plot, (c) Mott–Schottky plot, and (d) current density vs potential under dark and illuminated conditions (VRHE) of CuS and the heterojunction CuS/CdS.

where τ denotes the charge carrier lifetime and fm is the frequency of the maximum phase change. Evaluated τ values of CuS, CdS, and CuS/CdS were 8.4, 21, and 5.1 ms, respectively, at the frequencies of the maximum phase change. The lowest charge carrier lifetime observed for the heterojunction is referring to fast charge transfer.

The Nyquist plot represents the plot of real impedance vs imaginary impedance, which is best fitted to an equivalent circuit. Under dark conditions, the observed impedance spectra of the heterojunction CuS/CdS show the lower charge transfer and ohmic resistance than bare CuS. Charge transfer resistance for CuS, CdS, and CuS/CdS was evaluated by RCT = Rmax – RΩ, where RΩ is ohmic resistance and Rmax is the maximum impedance toward the x-axis, which is depicted in Table .

Table 1

Electrochemical Properties, i.e., Ohmic Resistance, Charge Transfer Resistance, Short-Circuit Current Density (Dark and Illumination), and the Open-Circuit Voltage of CuS/CdS, CuS, and CdS Thin Films

samples	R (Ω)	RCT (Ω)	Jsc (A) dark	Jsc (A) light	Jsc (A) difference	Voc (V) light
CuS	58.55	118.95	–0.081 μ	–0.15 μ	0.069 μ	0.47
CdS	77.89	5325.99	5.04 μ	5.33 μ	0.29 μ	0.18
CuS/CdS	3.09	35.13	0.18 m	0.71 m	0.53 m	0.53

Mott–Schottky (M–S) Plots

The Mott–Schottky plots of CuS and CuS/CdS were analyzed by potential impedance spectroscopy. The M–S plot, i.e., inverse-square capacitance vs the applied potentials, with the interface of the electrode/electrolyte is displayed in Figure c. The M–S plot was obtained by applying sinusoidal waves with a change in potentials and the observed respective inverse capacitance. While observing the charge transfer during the range of applied potentials at the electrode/electrolyte interface, the capacitance varies depending upon the structural and electronic properties of the system under study. The electrode/electrolyte interface creates a Schottky barrier of the potential at which the linear charge transfer has a linear response of which extrapolation toward the x-axis gives the flat band potential (VFB). The semiconductor and barrier capacitance is in series with the electrolyte under negligible impedance. Thus, the capacitance due to the contribution of semiconductors and electrolytes was neglected. The M–S plots shown were obtained using the following M–S relations:[42]in which εo (8.85 × 10–12 farad per meter)[43] and εs (CuS = ∼3 at 40 °C, CdS = 9.35 at 25 °C )[44] are the permittivity of free space and the dielectric constant of the semiconductor, A (0.5 cm–2 for all thin films) is the area of the working electrode under study in the PEC cell, k is Boltzmann’s constant, and e is the charge of an electron. The slope for the n-type semiconductor and for the p-type semiconductor was evaluated by extrapolating the M–S curve toward the x-axis. The evaluated donor concentration of CdS and acceptor concentration of CuS are 10.31 × 1016 and 30.79 × 1016, respectively, using M–S relations.

Current Density vs Potential (J–V) Plot

LSV was carried out for the heterojunction CuS/CdS, bare CuS, and CdS with an area 0.5 cm2, and the potentials were converted to RHE. Current density vs potential (RHE) plots of bare CuS and CuS/CdS are displayed in Figure d. The short-circuit current density (Jsc) and open-circuit voltage (Voc) are determined for CdS, CuS, and CuS/CdS, which are displayed in Table . The current density improved by ∼99.9% for the CuS/CdS heterojunction compared to bare CuS NPs. The photoconversion efficiencies evaluated for CuS and the CuS/CdS heterojunction are 1.7 × 10–5 (Fill Factor = 23.82%) and 0.65% (Fill Factor = 17.23%), respectively.

Cyclic Voltammetric Curves of CuS and CdS

For the determination of band edge positions, cyclic voltammetry experiments were performed for CuS and CdS thin films in the electrolyte 0.5 M Na2SO4 at room temperature using the standard three-electrode system shown in Scheme . Figure a,b shows the CV curves of CuS and CdS including anodic (a1) and cathodic (c1) peaks for the electrodes under study. The observed cathodic peaks for CuS and CdS are −1.22 V vs VSCE and −1 V vs VSCE, respectively. Similarly, anodic peaks at 1 V vs VSCE and 0.98 V vs VSCE are for CuS and CdS, respectively. The lower unoccupied molecular orbital (LUMO), higher occupied molecular orbitals (HOMO), and electrochemical band gaps were estimated using the following expressions:[45]

Figure 4

(a) CV curves of CuS and (b) CdS in which a1 and c1 are anodic and cathodic peaks, respectively.

Here, Eg is the electrochemical band gap, ERED is a cathodic peak (c1), EOX is an anodic peak (a1), and the potential of 4.68 eV represents the potential difference between the SCE electrode and an electron in vacuum. Evaluated parameters are listed in Table .

Table 2

Electrochemical, Optical Band Gap and Electron Affinity of CuS and CdS

sample	EVB/vacuum	ECB/vacuum	electrochemical band gap (eV)	optical band gap (eV)
CuS	5.90	3.68	2.22	2.10
CdS	5.68	3.70	1.98	2.00

CuS/CdS Interface

CV data were analyzed to understand the band alignment between CuS and CdS as p-type and n-type semiconductor heterojunctions. Valence band maxima (VBM), conduction band minima (CBm), and conduction and valence band offsets are evaluated by applying HOMO and LUMO expressions. Figure shows the band alignment of type-1 for the heterojunctions CuS and CdS. ΔEc is the difference between the electron affinities (χ), and ΔEv is the difference between the VBM of the semiconductors. ΔEc and ΔEv denote the conduction band offset (CBO) and the valence band offset (VBO), which are 0.02 and 0.24 eV, respectively. Under the influence of a negative potential and illumination, the electron transfers from the VBM of CuS to the CBm of CuS leaving the holes behind. Thereafter, the electron transfers from the CBm of CuS to the CBm of CdS while holes are moving in the reverse direction from the VBM of CdS to CuS. The observed narrow CBO is advantageous for reliable electron transfer, and the smaller VBO for the CuS/CdS interface leads to higher photogenerated charge carrier transfers.

Figure 5

Band diagram of the heterojunction CuS/CdS showing a straddling type-1 heterojunction in an electrochemical cell where platinum (Pt) is a counter electrode (CE) and a saturated calomel electrode (SCE) is the reference electrode with its potential value from the reference of vacuum, in the electrolyte of 0.5 M Na2SO4.

Mechanism of Photoelectrochemical Charge Transfer

In the photoelectrochemical cell, the heterojunction CuS/CdS shows the type-I heterojunction, resulting in a narrower CBO and VBO. In a straddling type-I heterojunction, both the charge carriers, electrons and holes, transfer toward a narrower band gap semiconductor as shown in Figure .[46] Under illumination, photogenerated holes and electrons accumulate at the valance band and the conduction band, respectively.[47] Due to the decreased redox potential differences, redox reactions occur at lower applied potentials and simultaneously.

Conclusions

In summary, we have illustrated a facile chemical route synthesis of pure CuS and CdS NPs. The electronic band structure using electrochemical analysis of the as-prepared materials was characterized by sophisticated experimental methods for the fabrication of a highly efficient photocathode. The fabricated straddling type-1 heterojunction photocathode CuS/CdS displays a huge enhancement in the photoconversion efficiency showing 0.65% compared to the pristine CuS photocathode with 1.7 × 10–5% at 0 VRHE. Reduced redox potentials and the improved absorption of light in the region of visible spectra resulted in the efficient CuS/CdS heterojunction. We believe that our approach for efficient and enhanced photoconversion and improved charge carrier transfer between the heterojunction and the electrode/electrolyte interface can be applied to other photocathodes to enhance the performance of the PEC cell.