The pursuit of efficient, low-cost, and sustainable photovoltaic technologies is a central driver in modern materials science. Among the various contenders, thin film solar panels offer distinct advantages such as reduced material usage, flexibility, and potential for large-area, low-temperature manufacturing. Within this domain, kesterite copper zinc tin sulfide (Cu2ZnSnS4 or CZTS) has emerged as a highly promising absorber material for the next generation of thin film solar panels. Its constituent elements are abundant and non-toxic, and its optical properties—a direct bandgap of approximately 1.5 eV and a high absorption coefficient exceeding 104 cm-1—are nearly ideal for efficient sunlight harvesting. The theoretical Shockley-Queisser efficiency limit for a CZTS-based thin film solar panel is around 32.4%. However, the champion laboratory cell efficiency currently stands at only 9.19%, revealing a significant performance gap. Key limitations include high charge carrier recombination losses, often linked to point defects and secondary phases, and poor film morphology with small grains and voids that hinder carrier transport and collection.
One strategic approach to mitigate these issues is cationic substitution. Partial replacement of copper (Cu) with silver (Ag) in the kesterite lattice, forming (Ag,Cu)2ZnSnS4 (ACZTS), has been proposed as a pathway to improve the material’s properties. Ag has a larger ionic radius and can form bonds with different strengths compared to Cu, which can influence defect formation energies, grain growth kinetics, and electronic band structure. This work investigates the effects of systematic Ag incorporation on the compositional, structural, morphological, and optoelectronic properties of ACZTS thin films fabricated via a two-step process, providing insights critical for the advancement of kesterite-based thin film solar panel technology.
1. Experimental Methodology for Thin Film Fabrication
The fabrication of the ACZTS absorber layers for thin film solar panel applications followed a sequential two-step process: electrochemical co-deposition of a metallic precursor and post-deposition reactive annealing. This method is attractive for its potential scalability and cost-effectiveness.
1.1 Precursor Deposition
The substrates were 20 mm × 20 mm soda-lime glass slides coated with an ~800 nm molybdenum (Mo) back contact layer. The metallic precursor stack was fabricated in two stages. First, a Cu-Zn-Sn (CZT) layer was electrodeposited onto the Mo substrate from an aqueous electrolyte bath. The bath composition and operating parameters were optimized to achieve a uniform and adherent deposit, as summarized in Table 1.
| Chemical | Concentration (g/L) | Role |
|---|---|---|
| CuSO4 | 1.60 | Cu source |
| SnSO4 | 2.15 | Sn source |
| ZnSO4·7H2O | 2.875 | Zn source |
| C6H8O7·H2O | 2.10 | Complexing agent & pH buffer |
| C6H5Na3O7·2H2O | 26.65 | Complexing agent & pH buffer |
Table 1: Composition of the electrochemical bath for CZT deposition. The bath pH was adjusted to 6.4. Deposition was carried out at a constant potential of -1.3 V (vs. Ag/AgCl) for 15 minutes.
Subsequently, a thin layer of silver was deposited onto the CZT precursor using DC magnetron sputtering. A high-purity Ag target (99.99%) was used under an Ar atmosphere. By varying the sputtering time from 0 to 4 minutes while keeping other parameters constant (base pressure: 4.5×10-4 Pa, working pressure: 5 Pa, power: 50 W), a series of precursors with different nominal Ag thicknesses, and hence different Ag/(Ag+Cu) atomic ratios, were prepared.
1.2 Sulfurization and Crystallization
The metallic precursor stacks were converted into crystalline sulfide thin films via a high-temperature annealing process. The samples were placed in a graphite box alongside 20 mg of sulfur powder. The box was then heated in a tube furnace under a flowing N2 atmosphere at a pressure of 4 kPa. The thermal profile involved ramping to 550°C at a controlled rate, holding at this temperature for 60 minutes to allow for complete sulfur incorporation and grain growth, and then cooling naturally. This process transforms the metal stack into the polycrystalline ACZTS absorber layer, a crucial step in fabricating the active component of a thin film solar panel.
1.3 Characterization Techniques
The properties of the resulting ACZTS thin films were extensively characterized to correlate Ag content with performance:
- Composition: Energy-dispersive X-ray spectroscopy (EDS) was used to determine the elemental atomic ratios.
- Structure & Phase: X-ray diffraction (XRD) and Raman spectroscopy were employed to identify crystalline phases and detect secondary compounds.
- Morphology: Scanning electron microscopy (SEM) was used to analyze surface topography and grain size.
- Optoelectronic Properties: The electronic and photoresponse characteristics were evaluated using electrochemical methods in a three-electrode cell with a 0.5 M Na2SO4 electrolyte. Mott-Schottky analysis was performed to deduce semiconductor type and flat-band potential. Photoelectrochemical (PEC) measurements under simulated AM1.5G illumination (100 mW/cm²) provided transient and steady-state photocurrent densities, which are proxy indicators for carrier generation and separation efficiency within the thin film solar panel absorber material.
2. Results and Discussion: The Impact of Silver
2.1 Compositional and Structural Evolution
The EDS-derived compositions of the sulfurized films are presented in Table 2. The parameter \( x = \text{Ag}/(\text{Ag}+\text{Cu}) \) defines the Ag atomic ratio relative to the total Group IB elements. As intended, \( x \) increased from 0 to 0.34 with longer Ag sputtering times. All films exhibited a Cu-poor and Zn-rich composition, a condition often targeted in high-efficiency kesterite thin film solar panels to suppress the formation of detrimental CuZn antisite defects. The (Ag+Cu)/(Zn+Sn) ratio was below 1, and Zn/Sn varied between 1.25 and 1.49.
| Ag Sputtering Time (min) | \( x = \text{Ag}/(\text{Ag+Cu}) \)** | (Ag+Cu)/(Zn+Sn) | Zn/Sn |
|---|---|---|---|
| 0 | 0.00 | 0.61 | 1.28 |
| 1 | 0.08 | 0.77 | 1.25 |
| 2 | 0.22 | 0.84 | 1.38 |
| 3 | 0.28 | 0.88 | 1.49 |
| 4 | 0.34 | 0.96 | 1.38 |
Table 2: Chemical composition of ACZTS thin films as a function of Ag sputtering time.
XRD patterns confirmed the formation of the kesterite structure. The primary diffraction peaks corresponding to the (112), (200), (220), and (312) planes of CZTS were observed. A clear shift in the dominant (112) peak position to lower diffraction angles (2θ) with increasing \( x \) was evident, as shown in the comparison below. This shift indicates an expansion of the lattice constant, which obeys Vegard’s law for a solid solution. The larger covalent radius of Ag⁺ (1.28 Å) compared to Cu⁺ (1.17 Å) causes lattice dilation when Ag substitutes for Cu, confirming the successful formation of the (Cu1-xAgx)2ZnSnS4 alloy, a key achievement for tailoring the absorber in a thin film solar panel.
$$ \Delta d_{112} \propto x \cdot (r_{\text{Ag}^+} – r_{\text{Cu}^+}) $$
Raman spectroscopy provided complementary phase identification, crucial for detecting secondary phases often invisible in XRD due to overlapping peaks or low crystallinity. All films showed the characteristic Raman modes of kesterite CZTS near 286 cm⁻¹ (A₁ mode) and 332 cm⁻¹. For films with \( x \geq 0.22 \), an additional broad feature emerged in the 250-270 cm⁻¹ range, which can be attributed to Ag₂S phases. This becomes more prominent at high Ag contents, indicating the likely presence of Ag₂S secondary phases alongside the ACZTS kesterite phase, a factor that can influence the electronic properties of the thin film solar panel absorber.
2.2 Dramatic Improvement in Thin Film Morphology
The surface morphology of the absorbers, a critical factor for device performance in thin film solar panels, was profoundly affected by Ag incorporation. SEM images revealed a striking evolution. The undoped CZTS film (\( x = 0 \)) exhibited a rough surface composed of small, poorly interconnected grains with numerous voids and pinholes. Such a morphology is detrimental as it increases series resistance, reduces the effective light-absorbing volume, and provides numerous recombination pathways at grain boundaries and surfaces.
With the introduction of Ag (\( x = 0.08 \)), a significant improvement was observed: grain size increased, and the surface became denser with fewer voids. This trend continued for \( x = 0.22 \) and \( x = 0.28 \), where the films displayed compact, densely packed, and larger grains, forming a high-quality surface coverage ideal for a thin film solar panel absorber layer. The mechanism is likely related to the formation of a low-temperature molten Ag-Cu-Sn-S phase during sulfurization. This liquid phase enhances atomic mobility and promotes grain growth via a liquid-phase assisted sintering mechanism, leading to superior microstructure. However, at the highest Ag content (\( x = 0.34 \)), the grain size slightly decreased, and large, distinct particulates appeared on the surface, consistent with the formation of Ag₂S segregations as suggested by Raman data. This highlights an optimal range for Ag incorporation to maximize microstructural benefits without excessive secondary phase formation.
2.3 Modification of Surface Electronic Properties
Mott-Schottky analysis was employed to investigate the semiconductor type and flat-band potential at the film/electrolyte interface, providing insight into the near-surface electronic structure relevant for junction formation in a thin film solar panel. The analysis is based on the relationship between the space-charge capacitance (\(C_{SC}\)) and the applied potential (\(E\)):
$$ \frac{1}{C_{SC}^2} = \frac{2}{\epsilon \epsilon_0 e N_d} \left( E – E_{fb} – \frac{k_B T}{e} \right) $$
where \( \epsilon \) is the dielectric constant, \( \epsilon_0 \) the vacuum permittivity, \( e \) the elementary charge, \( N_d \) the charge carrier density, \( E_{fb} \) the flat-band potential, \( k_B \) Boltzmann’s constant, and \( T \) the temperature. The slope of the \( 1/C^2 \) vs. \( E \) plot indicates the semiconductor type: a positive slope for n-type and a negative slope for p-type.
The undoped CZTS film exhibited a negative slope across the measured range, confirming its p-type character. For Ag-doped films (\( x \geq 0.08 \)), the Mott-Schottky plots showed more complex behavior. While sections of the curves retained a negative slope (p-type), distinct regions with a positive slope (n-type) also appeared. This bipolar character at the surface is attributed to the coexistence of the p-type ACZTS kesterite phase and n-type Ag₂S secondary phases detected by Raman spectroscopy. The presence of an n-type surface layer on a p-type absorber could potentially form a buried homojunction or modify band alignment, which could be engineered to benefit carrier collection in a thin film solar panel, though uncontrolled formation likely leads to recombination.
2.4 Enhanced Photoelectrochemical Response
The photoresponse of the ACZTS thin films, a direct indicator of their viability as an absorber material, was evaluated through chopped light chronoamperometry (I-t curves) at a fixed bias. Upon illumination, a sharp photocurrent spike (transient current, \(J_{trans}\) ) is generated due to the rapid separation of photogenerated electron-hole pairs. This current then decays to a steady-state value (\(J_{steady}\) ) as recombination processes equilibrate. The magnitudes of \(J_{trans}\) and \(J_{steady}\) are related to the initial carrier generation/separation efficiency and the steady-state balance between generation and recombination, respectively.
The photocurrent densities for all films are summarized in Table 3. A non-monotonic dependence on Ag content \( x \) is observed for both parameters.
| Ag Ratio (\( x \)) | Transient Photocurrent Density, \(J_{trans}\) (mA/cm²) | Steady-State Photocurrent Density, \(J_{steady}\) (mA/cm²) |
|---|---|---|
| 0.00 | 0.07 | 0.03 |
| 0.08 | 0.15 | 0.06 |
| 0.22 | 0.32 | 0.09 |
| 0.28 | 0.25 | 0.11 |
| 0.34 | 0.18 | 0.08 |
Table 3: Photoelectrochemical performance metrics of ACZTS thin films.
The peak in \(J_{trans}\) at \( x = 0.22 \) suggests an optimal point for initial carrier generation and separation. This enhancement can be linked to two factors modulated by Ag: (1) a potential reduction in the optical bandgap (\(E_g\)) of the ACZTS alloy within this composition range, allowing absorption of a broader spectrum of the solar irradiance, and (2) the significantly improved film density and reduced void fraction, which minimizes bulk recombination immediately upon photo-excitation. The bandgap tuning in ACZTS can be empirically modeled as a function of \( x \):
$$ E_g(x) \approx E_g(\text{CZTS}) – \alpha x + \beta x^2 $$
where \( \alpha \) and \( \beta \) are coefficients describing the initial bandgap bowing and subsequent increase, respectively, with \( \alpha > \beta \). The maximum \(J_{steady}\) at \( x = 0.28 \) underscores the importance of morphological perfection for sustained carrier collection. The dense, large-grained microstructure at this composition minimizes recombination centers (grain boundaries, voids) throughout the bulk of the thin film solar panel absorber, allowing more photogenerated carriers to reach the interface. The decline in performance at \( x = 0.34 \) correlates with the observed excessive Ag₂S formation and slight grain coarsening degradation, which introduce new recombination pathways and potentially create current-blocking barriers, negating the benefits of Ag doping.
3. Conclusions and Outlook for Thin Film Solar Panels
This investigation systematically demonstrates that silver incorporation is a powerful tool for engineering the properties of kesterite CZTS thin film absorbers. The synthesis of (Cu1-xAgx)2ZnSnS4\) solid solutions was confirmed through lattice expansion and phase analysis. Most notably, Ag acts as a potent microstructural modifier, transforming a rough, porous CZTS film into a dense, large-grained, and compact layer via a proposed liquid-phase assisted growth mechanism during sulfurization. This morphological improvement directly addresses a key bottleneck in current kesterite thin film solar panel technology.
The optoelectronic characterization reveals a complex interplay of effects. Ag doping modifies the surface electronic character, introducing n-type regions likely associated with Ag₂S, and significantly enhances the photoelectrochemical response. The transient photocurrent, linked to initial carrier yield, peaked at an Ag ratio of \( x = 0.22 \), while the steady-state photocurrent, indicative of effective carrier collection, was highest at \( x = 0.28 \). This indicates that different Ag concentrations optimize different aspects of performance: a moderate level may be best for maximizing light absorption and initial separation, while a slightly higher level is superior for minimizing bulk recombination through optimal grain structure.
These findings provide clear guidelines for the development of high-efficiency ACZTS-based thin film solar panels. The optimal Ag content lies in the range of \( x = 0.22 \) to 0.28, where the benefits of grain growth and densification are maximized without severe degradation from secondary phase segregation. Future work should focus on integrating these optimized absorbers into full photovoltaic device stacks, carefully managing the formation of secondary phases like Ag₂S, and exploring the impact of the Ag-induced surface electronic modification on the p-n junction properties. By leveraging the positive effects of Ag on both morphology and optoelectronic response, the pathway toward closing the efficiency gap for sustainable kesterite thin film solar panels becomes more tangible.