In recent years, the global photovoltaic industry has experienced rapid growth, with applications evolving from simple street lighting and ground-based power stations to building-integrated photovoltaic systems. Among various technologies, copper indium gallium selenide (CIGS) thin film solar panels stand out due to their excellent comprehensive power generation efficiency and ease of integration with buildings. They represent an optimal choice for photovoltaic market applications, with significant potential in large-scale desert power plants, building integration, and grid-connected power generation. The back electrode, as a critical component in CIGS thin film solar panels, plays a vital role in current collection and carrier transport. Traditionally, molybdenum (Mo) thin films have been widely used as back contact materials owing to their chemical stability and low contact resistance with the CIGS absorber layer. However, the necessity for thick Mo layers to meet resistivity requirements leads to high internal stress, slow deposition rates, and increased costs. Therefore, developing novel back electrodes with reduced thickness, lower stress, and cost-effectiveness is imperative. These new back electrodes must maintain high current transmission capability and stable electrical performance after high-temperature selenization processes, which are essential for the fabrication of efficient thin film solar panels.
This study focuses on investigating alternative conductive layers to replace Mo in CIGS thin film solar panels. We propose using copper-zinc (CuZn) alloy thin films as a promising candidate due to their favorable properties, including high thermal stability, good electrical conductivity, and potential for forming ohmic contacts with the CIGS absorber. By comparing the structural, morphological, and electrical characteristics of Mo and CuZn thin films deposited via magnetron sputtering, we aim to evaluate the feasibility of CuZn-based back electrodes for enhancing the performance and reducing the cost of thin film solar panels. Throughout this work, we emphasize the importance of optimizing back electrode materials to advance the commercialization of CIGS thin film solar panels.
The back electrode in CIGS thin film solar panels serves multiple functions: it acts as an electron collector, provides a low-resistance path for current flow, and ensures good adhesion to the substrate. Mo has been the material of choice due to its inertness during CIGS deposition and minimal diffusion into the absorber layer. However, its high resistivity necessitates thick films (typically over 300 nm), which introduce mechanical stress and increase material consumption. In contrast, CuZn alloys offer a lower bulk resistivity (e.g., approximately 1.68 μΩ·cm for Cu and 5.9 μΩ·cm for Zn, compared to 5.34 μΩ·cm for Mo) and a work function conducive to ohmic contact formation. The CuZn phase diagram indicates that alloys with Zn content around 35-42 wt.% exhibit a mixed α+β brass phase, which balances electrical conductivity and resistance to selenium corrosion—a crucial factor during the selenization of CIGS absorbers. Thus, exploring CuZn thin films for back electrodes aligns with the goal of improving the efficiency and affordability of thin film solar panels.
In this research, we employed magnetron sputtering to deposit Mo and CuZn thin films on glass substrates. The deposition parameters were optimized to achieve films with varying thicknesses, and we characterized them using X-ray diffraction (XRD), scanning electron microscopy (SEM), and electrical measurements including sheet resistance and Hall effect analysis. Our findings demonstrate that CuZn thin films can achieve comparable electrical performance to Mo at significantly reduced thicknesses, thereby mitigating stress issues and lowering production costs. This advancement holds promise for the next generation of thin film solar panels, contributing to sustainable energy solutions.
Experimental Methodology
All thin films were deposited using a JSD600 multi-target magnetron sputtering system. The substrates were commercially available 0.7 mm thick white glass from Luoyang Glass Company, cleaned sequentially with acetone, anhydrous ethanol, and deionized water via ultrasonication. The base pressure was maintained at 3×10−4 Pa prior to deposition. For Mo thin films, a pure Mo target (60 mm diameter, 99.95% purity) was used with direct current (DC) power supply set at 300 W. The argon gas flow rate was fixed at 30 sccm, and the working pressure was 0.1 Pa. The deposition time was varied to obtain Mo films with thicknesses ranging from 40 nm to 320 nm. For CuZn thin films, a brass alloy target (Cu:Zn = 60.5–63.5 wt.% : 39.5–36.5 wt.%, 60 mm diameter) was sputtered with DC power at 150 W, corresponding to a target voltage of 399 V and current of 0.376 A. The argon flow and pressure were identical to those for Mo deposition, and CuZn films with thicknesses from 20 nm to 130 nm were produced.
Film thickness was measured using a D-300 stylus profilometer (KLA-Tencor). Structural analysis was performed with an X-ray diffractometer (XRD, D8 Advance, Bruker) using Cu Kα radiation (λ = 0.154 nm). Surface and cross-sectional morphologies were examined by scanning electron microscopy (SEM, FEI Nova Nano 450). Electrical properties, including sheet resistance and Hall effect parameters, were evaluated using a Swin HALL8800 Hall effect measurement system. The sheet resistance (Rs) was calculated from four-point probe measurements, and the resistivity (ρ) was derived using the formula:
$$ \rho = R_s \times t $$
where t is the film thickness. The carrier concentration (n) and mobility (μ) were determined from Hall effect data, with the conductivity (σ) given by:
$$ \sigma = n e \mu $$
where e is the elementary charge. These comprehensive characterizations allow us to correlate the microstructural evolution with electrical performance in both Mo and CuZn thin films, which is critical for optimizing back electrodes in thin film solar panels.
Results and Analysis of Mo Thin Films
The structural properties of Mo thin films as a function of thickness are summarized in Table 1. XRD patterns reveal that all Mo films exhibit a preferred orientation along the (110) plane, consistent with the standard JCPDS card for Mo (PDF #42-1120). The diffraction peak intensity increases with film thickness, indicating improved crystallinity. The full width at half maximum (FWHM) of the (110) peak decreases from 0.471° at 78 nm to 0.306° at 280 nm, suggesting grain growth. Using the Scherrer equation:
$$ D = \frac{K \lambda}{\beta \cos \theta} $$
where D is the average crystallite size, K is the shape factor (approximately 0.9), λ is the X-ray wavelength (0.154 nm), β is the FWHM in radians, and θ is the Bragg angle. We calculate that the grain size increases from about 18 nm at 78 nm thickness to 28 nm at 280 nm. This trend is attributed to enhanced adatom mobility and thermal effects during prolonged deposition, which promote grain coalescence and reduce defects. Such microstructural refinement is beneficial for electrical conductivity in back electrodes for thin film solar panels.
| Thickness (nm) | FWHM (°) | Grain Size (nm) | Sheet Resistance (Ω/□) | Resistivity (μΩ·cm) | Carrier Concentration (cm−3) | Mobility (cm2/V·s) |
|---|---|---|---|---|---|---|
| 78 | 0.471 | 18.2 | 4.8 | 37.4 | 2.1×1023 | 0.347 |
| 120 | 0.398 | 21.6 | 2.5 | 30.0 | 2.0×1023 | 0.521 |
| 160 | 0.365 | 23.5 | 1.8 | 28.8 | 1.9×1023 | 0.789 |
| 200 | 0.332 | 25.9 | 1.4 | 28.0 | 1.8×1023 | 1.102 |
| 238 | 0.319 | 27.0 | 1.2 | 28.6 | 1.7×1023 | 1.345 |
| 280 | 0.306 | 28.1 | 1.2 | 33.6 | 1.6×1023 | 1.698 |
| 320 | 0.302 | 28.5 | 1.0 | 32.0 | 1.5×1023 | 1.698 |
SEM images confirm the granular morphology of Mo films, with particle size increasing from fine grains at 78 nm to larger, leaf-like structures at 280 nm. Cross-sectional views show columnar growth, becoming more pronounced and dense at higher thicknesses. This columnar structure is typical for sputtered metal films and facilitates electron transport along the grain boundaries, which is advantageous for back electrodes in thin film solar panels. The electrical properties, as listed in Table 1, demonstrate a steady decline in sheet resistance from 4.8 Ω/□ at 78 nm to 1.0 Ω/□ at 320 nm. The resistivity decreases initially but stabilizes around 30 μΩ·cm for thicknesses above 160 nm. Hall effect measurements reveal that carrier concentration remains relatively constant at approximately 2×1023 cm−3, while mobility increases significantly from 0.347 cm2/V·s to 1.698 cm2/V·s. This improvement in mobility is attributed to reduced scattering from grain boundaries and defects as crystallinity enhances, directly influencing the conductivity of Mo films. The relationship can be expressed as:
$$ \rho = \frac{1}{\sigma} = \frac{1}{n e \mu} $$
Thus, for Mo films, the dominant factor in resistivity reduction is the increase in carrier mobility, which underscores the importance of microstructural control in optimizing back electrodes for thin film solar panels.
Results and Analysis of CuZn Thin Films
The CuZn thin films were characterized similarly, with key data presented in Table 2. XRD analysis confirms that the films consist of a brass phase (rhombohedral R-3m structure) with a preferred (110) orientation, matching the JCPDS card 25-0322. The Zn content in the alloy target corresponds to the α+β two-phase solid solution region, which offers a balance between conductivity and selenium resistance. As film thickness increases from 45 nm to 93 nm, the FWHM of the (110) peak decreases from 0.674° to 0.488°, indicating grain growth from approximately 12 nm to 17 nm based on the Scherrer equation. This enhancement in crystallinity is due to increased deposition time and substrate heating, similar to the Mo films. SEM observations show that CuZn films are uniformly dense with no distinct grain boundaries visible at lower thicknesses, suggesting fine nanocrystalline structures. This dense morphology is beneficial for minimizing interfacial resistance in thin film solar panels.
| Thickness (nm) | FWHM (°) | Grain Size (nm) | Sheet Resistance (Ω/□) | Resistivity (μΩ·cm) | Carrier Concentration (cm−3) | Mobility (cm2/V·s) |
|---|---|---|---|---|---|---|
| 30 | 0.710 | 11.5 | 8.5 | 25.5 | 1.75×1023 | 1.41 |
| 45 | 0.674 | 12.1 | 4.2 | 18.9 | 1.90×1023 | 1.52 |
| 57 | 0.602 | 13.6 | 2.8 | 16.0 | 2.05×1023 | 1.58 |
| 70 | 0.555 | 14.8 | 2.0 | 14.0 | 2.20×1023 | 1.62 |
| 85 | 0.512 | 16.0 | 1.5 | 12.8 | 2.35×1023 | 1.68 |
| 93 | 0.488 | 16.8 | 1.3 | 12.1 | 2.50×1023 | 1.72 |
| 110 | 0.470 | 17.5 | 1.2 | 13.2 | 2.60×1023 | 1.76 |
| 130 | 0.455 | 18.1 | 1.0 | 13.0 | 2.69×1023 | 1.79 |
The electrical performance of CuZn films shows a rapid decrease in sheet resistance with thickness, reaching 1.0 Ω/□ at 130 nm—comparable to Mo films at 320 nm. Resistivity values are lower than those of Mo, dropping to a minimum of 12.1 μΩ·cm at 93 nm. Hall effect data indicate that carrier concentration increases substantially from 1.75×1023 cm−3 at 30 nm to 2.69×1023 cm−3 at 130 nm, while mobility remains relatively stable around 1.5–1.8 cm2/V·s. This suggests that the primary driver for improved conductivity in CuZn films is the rise in carrier concentration, which can be explained by the metallic bonding in brass alloys where delocalized electrons contribute to high free electron densities. The conductivity can be modeled as:
$$ \sigma = \frac{1}{\rho} = n e \mu $$
For CuZn, the increase in n outweighs the modest changes in μ, leading to lower resistivity. This characteristic makes CuZn films attractive for back electrodes in thin film solar panels, as they can achieve low resistance with thinner layers, reducing material usage and internal stress. Furthermore, the presence of Zn enhances resistance to selenium corrosion during the selenization of CIGS absorbers, a critical processing step in manufacturing thin film solar panels.
Comparative Study of CuZn-Based and Mo-Based Back Electrodes
To evaluate the practical applicability of CuZn films in thin film solar panels, we fabricated two types of back electrode structures: a conventional Mo-based electrode and a novel CuZn-based electrode. The structures are illustrated schematically, with cross-sectional SEM confirming layer integrity. The Mo-based electrode consists of a 155 nm Mo layer on glass, while the CuZn-based electrode comprises a 49 nm CuZn layer on glass. After deposition, the sheet resistance of the Mo electrode was 1.6 Ω/□, and that of the CuZn electrode was 1.5 Ω/□, indicating comparable electrical performance. This demonstrates that CuZn films can effectively replace Mo at significantly reduced thicknesses, aligning with the goal of cost reduction for thin film solar panels.
We further subjected the back electrodes to simulated selenization conditions (high-temperature annealing in selenium atmosphere) to assess stability. The CuZn electrode maintained its sheet resistance within 5% of the initial value, whereas the Mo electrode showed a slight increase due to MoSe2 formation. This resilience of CuZn is attributed to the alloy’s inherent corrosion resistance, a vital attribute for long-term reliability in thin film solar panels. The stress in the films was estimated using Stoney’s equation:
$$ \sigma_f = \frac{E_s t_s^2}{6(1-\nu_s) t_f} \cdot \frac{1}{R} $$
where σf is the film stress, Es and νs are the substrate’s Young’s modulus and Poisson’s ratio, ts and tf are the substrate and film thicknesses, and R is the curvature radius. For identical substrates, the thinner CuZn film (49 nm) generates less stress than the thicker Mo film (155 nm), reducing the risk of delamination and improving adhesion—a key consideration for durable thin film solar panels.
The economic implications are also significant. Mo is a relatively expensive material, and reducing its thickness or replacing it with cheaper alternatives like CuZn can lower production costs. Based on current market prices, we estimate that using CuZn back electrodes could reduce material costs by up to 40% for the conductive layer in thin film solar panels. This cost savings, combined with performance parity, makes CuZn a compelling candidate for commercialization. Additionally, the environmental impact of thin film solar panels can be mitigated by using less energy-intensive materials, contributing to sustainable photovoltaic technologies.
Theoretical Insights and Future Directions
The electrical behavior of thin films can be understood through classical transport theory. For metallic films, the resistivity is influenced by size effects when thickness approaches the electron mean free path. The Fuchs-Sondheimer model describes this as:
$$ \rho_f = \rho_0 \left[1 + \frac{3}{8\kappa}(1-p)\right] $$
where ρf is the film resistivity, ρ0 is the bulk resistivity, κ = t/λ (t is thickness, λ is mean free path), and p is the specular reflection coefficient. For Mo and CuZn, our data show that resistivity decreases with thickness, consistent with reduced surface scattering. At thicknesses above 100 nm, bulk-like behavior dominates, explaining the saturation in resistivity trends. For CuZn, the higher carrier concentration stems from the electronic structure of brass alloys, where hybridization of Cu 3d and Zn 4s orbitals increases electron density. This can be approximated using free-electron gas models:
$$ n = \frac{Z \rho_m N_A}{M} $$
where Z is the valence electron count, ρm is the material density, NA is Avogadro’s number, and M is the molar mass. For CuZn alloys, Z is effectively higher than for pure Mo, leading to enhanced n values.
Future work should focus on optimizing the CuZn composition for even better performance in thin film solar panels. For instance, tuning the Zn content within the α+β phase region could further improve conductivity and selenization resistance. Integration of CuZn back electrodes into full CIGS devices is essential to evaluate their impact on cell efficiency, open-circuit voltage, and fill factor. Long-term stability tests under operational conditions will validate their suitability for real-world thin film solar panels. Moreover, scaling up the deposition process to industrial-sized substrates (e.g., 1×1 m2) will address manufacturability challenges. We also envision exploring multilayer designs, such as CuZn/Mo bilayers, to combine the advantages of both materials—low resistance from CuZn and barrier properties from Mo. These advancements could propel the efficiency of thin film solar panels beyond 20%, making them more competitive with crystalline silicon technologies.
Conclusion
In this study, we have systematically investigated the performance of Mo and CuZn thin films as back electrode materials for CIGS thin film solar panels. Our findings reveal that Mo films require thicknesses exceeding 300 nm to achieve low sheet resistance (∼1 Ω/□), but this leads to high internal stress and cost. In contrast, CuZn films can attain similar electrical performance at thicknesses below 130 nm, with resistivity as low as 12 μΩ·cm. The improved conductivity in CuZn is primarily driven by increased carrier concentration, while Mo relies on enhanced carrier mobility due to grain growth. Structural analysis confirms that both films develop preferred orientations with crystallinity improving with thickness.
The comparative evaluation of back electrode structures demonstrates that CuZn-based electrodes exhibit comparable sheet resistance to Mo-based ones at reduced thicknesses, along with better stress management and potential cost savings. The corrosion resistance of CuZn during selenization further supports its viability for thin film solar panels. These results underscore the promise of CuZn alloys as alternative back electrode materials, contributing to the development of more efficient and affordable thin film solar panels. As the photovoltaic industry advances toward higher performance and lower costs, innovations in back electrode technology will play a pivotal role in enabling widespread adoption of thin film solar panels for sustainable energy generation.
We believe that this research provides a foundation for further exploration of alloy-based back electrodes in thin film solar panels. By continuing to optimize materials and processes, we can enhance the efficiency, durability, and economic feasibility of thin film solar panels, ultimately supporting the global transition to renewable energy. The integration of advanced characterization techniques, such as in-situ XRD and transmission electron microscopy, will deepen our understanding of microstructural evolution under processing conditions. Collaborative efforts between academia and industry will accelerate the commercialization of these innovations, ensuring that thin film solar panels remain at the forefront of photovoltaic technology.