In the pursuit of enhancing the efficiency and applicability of renewable energy technologies, I have focused on the simulation and optimization of thin film solar panels, specifically those utilizing tandem junction structures. The primary goal of this research is to broaden the effective utilization of solar energy by maximizing the conversion of photons into electrical power, thereby improving the overall photoelectric conversion efficiency of the device. Traditional single-junction thin film solar panels, such as those based solely on amorphous silicon (a-Si), face limitations due to their inability to fully harness the broad solar spectrum. Photons with energies below the semiconductor’s bandgap do not contribute to current generation, while those with excessively high energies lead to thermal losses, capping the theoretical efficiency at around 14%. To overcome these constraints, I have employed a tandem cell approach, where multiple semiconductor layers with different bandgaps are stacked to capture a wider range of wavelengths. This study centers on a dual-junction configuration comprising an amorphous silicon top cell and an amorphous silicon-germanium (a-SiGe) bottom cell, which I simulate using advanced numerical methods to identify optimal design parameters for high-performance thin film solar panels.
The tandem cell structure fundamentally operates by serially connecting two pin junctions, each tailored to absorb specific segments of the solar spectrum. In an ideal scenario, the total photovoltage of the tandem thin film solar panel equals the sum of the individual sub-cell photovoltages, while the photocurrent is constrained by the smaller of the two sub-cell currents due to the continuity principle. Thus, meticulous material matching and layer thickness optimization are critical to balance current generation and minimize losses. For this investigation, I selected a wide-bandgap a-Si (approximately 1.72 eV) for the top cell to absorb high-energy photons in the visible range, and a narrower-bandgap a-SiGe alloy (around 1.50 eV) for the bottom cell to capture lower-energy near-infrared photons. This strategic pairing aims to reduce the Staebler-Wronski (S-W) effect, enhancing the stability of the thin film solar panel, while pushing efficiency beyond the limits of single-junction designs.
To model the electrical behavior and performance of this tandem thin film solar panel, I utilized the AMPS-1D (Analysis of Microelectronic and Photonic Structures) software, a powerful tool for simulating carrier transport in solid-state devices. AMPS-1D solves the one-dimensional Poisson’s equation coupled with electron and hole continuity equations under given boundary conditions, enabling the calculation of key parameters such as vacuum energy levels, quasi-Fermi levels, and current densities. The core equations governing the simulation are as follows. Poisson’s equation describes the electrostatic potential distribution: $$ \frac{d}{dx} \left( \varepsilon(x) \frac{d\Psi}{dx} \right) = q \left[ p(x) – n(x) + N_D^+(x) – N_A^-(x) + p_t(x) – n_t(x) \right] $$ where $\Psi$ is the electrostatic potential per unit positive charge at position $x$, $\varepsilon$ is the dielectric constant, $q$ is the electron charge, $p(x)$ and $n(x)$ are free hole and electron concentrations, $N_D^+$ and $N_A^-$ are ionized donor and acceptor concentrations, and $p_t(x)$ and $n_t(x)$ are trapped hole and electron concentrations. The electron and hole current densities are derived from: $$ J_n(x) = q \mu_n n(x) \frac{dE_{Fn}}{dx} $$ $$ J_p(x) = q \mu_p p(x) \frac{dE_{Fp}}{dx} $$ with $\mu_n$ and $\mu_p$ representing carrier mobilities, and $E_{Fn}$ and $E_{Fp}$ denoting the quasi-Fermi levels for electrons and holes. The continuity equations account for generation and recombination: $$ \frac{1}{q} \frac{dJ_n}{dx} = -G_{OP}(x) + R(x) $$ $$ \frac{1}{q} \frac{dJ_p}{dx} = -G_{OP}(x) – R(x) $$ where $G_{OP}(x)$ is the optical generation rate and $R(x)$ is the recombination rate. By solving these equations numerically, AMPS-1D provides insights into the performance metrics of the thin film solar panel, including short-circuit current ($J_{sc}$), open-circuit voltage ($V_{oc}$), fill factor (FF), and conversion efficiency ($\eta$).
For the simulation, I constructed a detailed physical model of the tandem thin film solar panel with the layer sequence: ITO (front contact) / p-type a-Si:H / intrinsic a-Si:H / n-type a-Si:H / p-type a-Si:H / intrinsic a-SiGe / n-type a-Si:H / ZnO / Ag (back contact). This structure is deposited on a stainless steel substrate, as illustrated in the schematic. The top cell’s intrinsic layer (i-a-Si:H) thickness was varied from 100 nm to 600 nm, while the bottom cell’s intrinsic layer (i-a-SiGe) thickness ranged from 1500 nm to 2500 nm, allowing me to explore the impact of layer dimensions on overall efficiency. The simulation assumed standard AM1.5 illumination with a power density of 100 mW/cm² at 300 K, ideal anti-reflection conditions at the front electrode (reflectivity = 0), and perfect reflection at the back electrode (reflectivity = 1). Carrier recombination velocities at both contacts were set to $1 \times 10^7$ cm/s to approximate realistic interfaces. The key material parameters for each layer in the thin film solar panel are summarized in Table 1, which includes bandgap, dielectric constant, carrier mobilities, doping concentrations, and density of states. These parameters were carefully selected based on experimental data to ensure the simulation accurately reflects the behavior of actual amorphous silicon and silicon-germanium alloys in thin film solar panels.
| Parameter | p-a-Si:H (Top) | i-a-Si:H (Top) | n-a-Si:H (Top) | p-a-Si:H (Bottom) | i-a-SiGe (Bottom) | n-a-Si:H (Bottom) |
|---|---|---|---|---|---|---|
| Thickness (nm) | 8 | Variable (100-600) | 8 | 8 | Variable (1500-2500) | 8 |
| Dielectric Constant | 11.9 | 11.9 | 11.9 | 11.9 | 11.9 | 11.9 |
| Bandgap (eV) | 1.8 | 1.72 | 1.72 | 1.70 | 1.50 | 1.6 |
| Electron Affinity (eV) | 3.9 | 3.9 | 3.9 | 4.05 | 4.05 | 4.05 |
| Electron Mobility (cm²V⁻¹s⁻¹) | 50 | 50 | 50 | 50 | 200 | 50 |
| Hole Mobility (cm²V⁻¹s⁻¹) | 5 | 5 | 5 | 5 | 100 | 5 |
| Doping Concentration (cm⁻³) | N_A = 1×10¹⁹ | N/A | N_D = 3×10¹⁹ | N_A = 1×10¹⁹ | N/A | N_D = 3×10¹⁹ |
| Effective Density of States in Conduction Band (cm⁻³) | 2.5×10²⁰ | 2.5×10²⁰ | 2.5×10²⁰ | 2.5×10²⁰ | 3×10¹⁹ | 2.5×10²⁰ |
| Effective Density of States in Valence Band (cm⁻³) | 2.5×10²⁰ | 2.5×10²⁰ | 2.5×10²⁰ | 2.5×10²⁰ | 2×10¹⁹ | 2.5×10²⁰ |
| Acceptor Activation Energy (eV above V.B.) | 0.25 | 0 | 0 | 0.25 | 0 | 0 |
| Donor Activation Energy (eV below C.B.) | 0 | 0 | 0.1 | 0 | 0 | 0.1 |
The simulation results revealed significant dependencies of the tandem thin film solar panel’s performance on the intrinsic layer thicknesses. By systematically varying the top and bottom cell thicknesses, I generated a comprehensive dataset to analyze efficiency trends. The conversion efficiency $\eta$ is calculated as: $$ \eta = \frac{J_{sc} \times V_{oc} \times FF}{P_{in}} \times 100\% $$ where $P_{in}$ is the incident power density (100 mW/cm²). For each configuration, I extracted $J_{sc}$, $V_{oc}$, and FF from AMPS-1D outputs. As shown in Table 2, which summarizes key results for selected thickness combinations, the efficiency peaks at a specific pair of values. I observed that when the bottom cell’s intrinsic layer (i-a-SiGe) is fixed at 2000 nm, varying the top cell’s intrinsic layer (i-a-Si:H) from 100 nm to 600 nm initially increases efficiency due to enhanced photon absorption in the visible range, but beyond 300 nm, efficiency declines because of increased recombination and reduced fill factor. Similarly, with the top cell fixed at 300 nm, adjusting the bottom cell thickness from 1500 nm to 2500 nm shows an optimal point at 2000 nm, where the trade-off between current generation and voltage loss is balanced. The highest efficiency achieved in the simulation was 14.46% for a configuration with 300 nm i-a-Si:H and 2000 nm i-a-SiGe, underscoring the importance of thickness optimization in tandem thin film solar panels.
| Top Cell i-layer Thickness (nm) | Bottom Cell i-layer Thickness (nm) | Short-Circuit Current Density, $J_{sc}$ (mA/cm²) | Open-Circuit Voltage, $V_{oc}$ (V) | Fill Factor, FF (%) | Conversion Efficiency, $\eta$ (%) |
|---|---|---|---|---|---|
| 100 | 2000 | 9.85 | 1.65 | 72.3 | 11.76 |
| 200 | 2000 | 11.20 | 1.68 | 73.1 | 13.78 |
| 300 | 2000 | 11.92 | 1.70 | 73.5 | 14.46 |
| 400 | 2000 | 12.05 | 1.69 | 72.8 | 14.12 |
| 500 | 2000 | 12.10 | 1.67 | 71.9 | 13.65 |
| 600 | 2000 | 12.15 | 1.65 | 70.5 | 12.98 |
| 300 | 1500 | 10.45 | 1.72 | 74.0 | 13.32 |
| 300 | 2000 | 11.92 | 1.70 | 73.5 | 14.46 |
| 300 | 2500 | 12.30 | 1.68 | 72.0 | 13.92 |
To visualize the structural configuration of such an optimized device, consider the following representation of a typical thin film solar panel with tandem junctions. This image highlights the layered architecture that enables efficient photon harvesting across the solar spectrum.
The efficiency trends can be further analyzed through mathematical modeling of the photocurrent generation. For a tandem thin film solar panel, the current matching condition is crucial, as the overall photocurrent $J_{total}$ is limited by the sub-cell with the lower current: $$ J_{total} = \min(J_{top}, J_{bottom}) $$ where $J_{top}$ and $J_{bottom}$ are the photocurrents of the top and bottom cells, respectively. These currents depend on the absorption profiles, which I calculated using the Beer-Lambert law: $$ G_{OP}(x) = \int_{\lambda} \Phi(\lambda) \alpha(\lambda) e^{-\alpha(\lambda) x} d\lambda $$ Here, $\Phi(\lambda)$ is the photon flux from the AM1.5 spectrum, and $\alpha(\lambda)$ is the absorption coefficient, which varies with material bandgap. For the a-Si top cell, $\alpha$ is higher for shorter wavelengths, while for the a-SiGe bottom cell, it extends into the infrared. The absorption coefficients can be approximated as: $$ \alpha(\lambda) = A \sqrt{\frac{hc/\lambda – E_g}{E_g}} $$ for photons with energy above the bandgap $E_g$, where $A$ is a material constant, $h$ is Planck’s constant, and $c$ is the speed of light. By integrating $G_{OP}(x)$ over the intrinsic layer thicknesses $d_{top}$ and $d_{bottom}$, I estimated the generated carrier densities and subsequently the photocurrents. The simulation in AMPS-1D incorporates these effects rigorously, accounting for drift-diffusion transport and recombination mechanisms such as Shockley-Read-Hall (SRH) recombination, which is prevalent in amorphous thin film solar panels. The recombination rate $R(x)$ in the continuity equations is given by: $$ R(x) = \frac{np – n_i^2}{\tau_p (n + n_t) + \tau_n (p + p_t)} $$ where $n_i$ is the intrinsic carrier concentration, and $\tau_n$ and $\tau_p$ are carrier lifetimes. In amorphous materials, these lifetimes are typically short due to high defect densities, underscoring the need for optimized layer thicknesses to minimize bulk recombination while maximizing absorption.
My simulation also involved sensitivity analysis on other parameters, such as doping profiles and interface properties, to assess their impact on the thin film solar panel performance. For instance, varying the p-layer doping concentration from $1 \times 10^{18}$ cm⁻³ to $1 \times 10^{20}$ cm⁻³ showed that higher doping improves $V_{oc}$ but can reduce $J_{sc}$ due to increased absorption in the doped layers, highlighting a trade-off. Additionally, I explored the effect of bandgap grading in the i-a-SiGe layer to enhance carrier collection, modeled by a linearly varying bandgap from 1.55 eV at the p-i interface to 1.45 eV at the n-i interface. This grading introduces an internal electric field that aids hole transport, potentially boosting the fill factor of the thin film solar panel. The modified bandgap profile is expressed as: $$ E_g(x) = E_{g0} + \frac{dE_g}{dx} \cdot x $$ where $E_{g0}$ is the bandgap at the interface and $\frac{dE_g}{dx}$ is the grading coefficient. Incorporating this into AMPS-1D required adjusting the electron affinity accordingly to maintain consistent band offsets.
Furthermore, I investigated the role of light trapping schemes, such as textured interfaces and back reflectors, which are common in advanced thin film solar panels to increase the effective optical path length. While my baseline simulation assumed ideal anti-reflection, I added a simple model for a textured front surface by modifying the generation rate to include a scattering factor $S$: $$ G_{OP,textured}(x) = G_{OP}(x) \times (1 + S e^{-\beta x}) $$ where $\beta$ is an attenuation coefficient. This enhancement led to a marginal efficiency improvement of about 0.5%, emphasizing that optical management is a complementary strategy to thickness optimization in tandem thin film solar panels. However, for this study, I focused primarily on the intrinsic layer thicknesses as the most critical design variables.
The results from this simulation have profound implications for the fabrication of high-efficiency thin film solar panels. By identifying 300 nm for the a-Si top cell and 2000 nm for the a-SiGe bottom cell as optimal, I provide a clear guideline for experimentalists to minimize trial-and-error in deposition processes. This configuration not only maximizes efficiency but also considers stability, as thinner a-Si layers reduce light-induced degradation. Moreover, the use of a-SiGe alloys with tailored bandgaps around 1.50 eV ensures better infrared response compared to pure a-Si, making the tandem thin film solar panel more versatile under real-world illumination conditions. My analysis also suggests that further gains could be achieved by incorporating microcrystalline silicon (µc-Si) as a bottom cell material, which offers higher carrier mobilities and stability, though that lies beyond the scope of this study.
In conclusion, through detailed numerical simulation using AMPS-1D, I have demonstrated the potential of amorphous silicon/amorphous silicon-germanium tandem junctions to achieve efficiencies exceeding 14% in thin film solar panels. The optimization of intrinsic layer thicknesses is pivotal, with a combination of 300 nm for the top cell and 2000 nm for the bottom cell yielding the highest performance. This research underscores the importance of computational modeling in advancing photovoltaic technology, offering a pathway to more efficient and stable thin film solar panels for widespread renewable energy adoption. Future work could extend this approach to triple-junction designs or explore novel materials like perovskites to push the boundaries of thin film solar panel efficiency even further.