In short, spectral irradiance—the distribution of power across different wavelengths of sunlight—directly and significantly impacts the energy output of all photovoltaic (PV) modules, but the magnitude and nature of this impact vary dramatically depending on the semiconductor material used in the cell. A module’s performance isn’t just about the total amount of light; it’s about how well that light’s “color recipe” matches the module’s specific “appetite” for certain wavelengths. This is why two modules with the same nameplate power rating under Standard Test Conditions (STC) can produce vastly different amounts of electricity in the real world, depending on the local atmospheric conditions, time of day, and season, all of which alter the solar spectrum.
To understand this, we first need to grasp the concept of the Air Mass (AM). This is a measure of the path length sunlight takes through the atmosphere relative to the path length when the sun is directly overhead (AM1.0). STC uses a spectrum defined for AM1.5, which represents a sun at about 48 degrees from the zenith, a typical average for mid-latitudes. However, the spectrum changes throughout the day. In the early morning and late afternoon, the sun’s rays travel through more atmosphere (e.g., AM2 or higher), which scatters shorter wavelengths (blue light) more effectively, resulting in a spectrum richer in longer wavelengths (red and infrared light). Conversely, at solar noon on a clear day, the spectrum is more balanced.
The key metric for quantifying a module’s response to these spectral shifts is the spectral response (SR). This is a graph that shows how efficiently a cell converts photons of different wavelengths into electrons. A cell cannot use photons with energy less than its bandgap (the material’s fundamental energy threshold), and photons with energy much higher than the bandgap lose their excess energy as heat. The “ideal” SR is a square function, but real-world physics makes it a curve. The critical link to performance is the useful fraction, which is the proportion of the incoming spectral irradiance that overlaps with the module’s spectral response curve.
The Impact on Specific PV Technologies
Different technologies have distinct bandgaps and SR curves, leading to unique spectral performance characteristics.
Crystalline Silicon (c-Si) – The Market Dominator
Crystalline silicon, which commands over 95% of the market, has a bandgap of approximately 1.1 eV. This means its spectral response is strongest in the near-infrared region, around 1100 nm, but it also has good response across the visible light spectrum. Because of this broad response, c-Si is often considered less sensitive to spectral variations than some other technologies. However, its performance is not immune. Under “red-rich” conditions like early mornings, late afternoons, or hazy days, c-Si modules can perform slightly better than expected relative to their STC rating. Conversely, under “blue-rich” conditions, such as at high altitudes or on very clear days near solar noon, their performance might be slightly lower. The annual energy impact of spectral variations for c-Si is typically in the range of 1-3%, but this can be higher in specific climates.
Thin-Film Cadmium Telluride (CdTe) – The Infrared Performer
CdTe has a nearly ideal bandgap for single-junction solar cells (~1.45 eV), which is higher than silicon’s. This gives it a spectral response that is shifted towards shorter wavelengths compared to c-Si; it peaks in the visible range and cuts off more sharply in the infrared. Consequently, CdTe modules are particularly sensitive to spectral changes. They tend to underperform in red-rich environments (high air mass) because a significant portion of the incident light is in the infrared, which CdTe cannot utilize. However, they can overperform in blue-rich environments (low air mass, clear skies). This makes their performance highly location-dependent. In regions like the southwestern United States with high direct normal irradiance and clear skies, CdTe’s spectral gains can contribute significantly to its competitive energy yield.
Thin-Film Copper Indium Gallium Selenide (CIGS) – The Versatile Player
The performance of CIGS is more complex because its bandgap can be tuned by varying the ratio of Gallium to Indium. A higher Gallium content increases the bandgap. Generally, CIGS modules have a spectral response that is broader than CdTe’s but different from c-Si’s, often with a very high quantum efficiency in the visible range. Their spectral behavior is therefore a middle ground, but they can be engineered to better match specific local spectral conditions. For instance, a CIGS formulation with a slightly higher bandgap might be optimized for sunnier, blue-rich climates.
Gallium Arsenide (GaAs) – The High-Efficiency Benchmark
Primarily used in concentrator and space applications due to cost, GaAs has a bandgap of ~1.43 eV, similar to CdTe. It exhibits an exceptionally high and sharp spectral response. Its performance is extremely sensitive to the exact spectral match, making it crucial for high-concentration systems to use precise solar tracking to maintain a consistent spectrum on the cell.
The following table provides a comparative summary of the spectral performance characteristics.
| PV Technology | Typical Bandgap (eV) | Spectral Response Peak | Performance in Red-Rich (High AM) Spectrum | Performance in Blue-Rich (Low AM) Spectrum |
|---|---|---|---|---|
| Crystalline Silicon (c-Si) | ~1.1 | ~1100 nm (Near-IR) | Slight Gain (1-3%) | Slight Loss (1-3%) |
| Cadmium Telluride (CdTe) | ~1.45 | ~500-800 nm (Visible) | Significant Loss (can be >5%) | Significant Gain (can be >5%) |
| Copper Indium Gallium Selenide (CIGS) | ~1.0 – 1.7 (tunable) | Varies with composition | Moderate Loss/Gain (depends on formulation) | Moderate Gain/Loss (depends on formulation) |
| Gallium Arsenide (GaAs) | ~1.43 | ~870 nm (Near-IR/Visible) | Significant Loss | Significant Gain |
Quantifying the Effect: Beyond Standard Test Conditions
The standard metric of efficiency at STC (25°C, 1000 W/m², AM1.5 spectrum) is insufficient for predicting real-world energy harvest. This is where the Performance Ratio (PR) and spectral correction factors come into play. The PR is the ratio of the actual energy output to the theoretically possible output if the system always operated at STC. Spectral mismatches are a key factor that causes PR to deviate from 100%.
Sophisticated energy prediction models, like those in the PVsyst software, incorporate spectral loss functions. These models use the average air mass for a location to estimate the typical spectral deviation from the AM1.5 standard and apply a correction factor to the predicted energy yield. For a project developer, ignoring these spectral effects can lead to a miscalculation of the Levelized Cost of Energy (LCOE) by several percentage points.
For example, a study comparing c-Si and CdTe installations in two locations—Albuquerque, New Mexico (high altitude, clear skies, blue-rich spectrum) and Tampa, Florida (lower altitude, more humid, more aerosol scattering)—found that the CdTe system in Albuquerque outperformed its STC-based prediction by over 4% annually due to spectral gains. The same CdTe technology in Tampa showed almost no spectral gain and even a slight loss at certain times of the year. The c-Si systems in both locations showed much smaller variations, within ±1.5%.
Practical Implications for System Design and Energy Forecasting
Understanding spectral effects is not just academic; it has direct consequences for financial models and technology selection.
1. Technology Selection for a Specific Site: When choosing between technologies for a large-scale solar farm, the local typical spectrum should be a consideration. If historical weather data shows a prevalence of clear skies and low air mass, a technology like CdTe might have an energy yield advantage over c-Si that is not apparent from their STC ratings alone. Conversely, in a climate often characterized by hazy or overcast conditions (which create a more diffuse, bluer light), the advantage might diminish or reverse. It’s a critical part of the due diligence process to model the expected energy yield using typical meteorological year (TMY) data that includes spectral information.
2. Accurate Energy Yield Modeling: For engineers and financiers, accurate energy yield predictions are paramount. Using simple models that only consider total irradiance and temperature will introduce error. High-fidelity modeling that accounts for spectral effects, along with soiling, degradation, and other loss factors, is essential for reducing project risk and securing financing. This is especially true for technologies known to have higher spectral sensitivity.
3. Interpretation of Monitoring Data: System operators monitoring the performance of their PV plants might see daily and seasonal performance fluctuations that cannot be explained by irradiance and temperature alone. A dip in performance ratio on a clear afternoon compared to a hazy one, even at similar irradiance levels, can often be attributed to spectral changes. Knowing this prevents false alarms about potential system faults.
When evaluating different options, it’s crucial to look at the complete picture of how a specific pv module will interact with the unique light conditions of your location. The choice between technologies involves a complex trade-off between cost, degradation rates, temperature coefficients, and spectral response. For instance, while a module might have a fantastic temperature coefficient (losing less power on hot days), its spectral response might be a poor match for a site that experiences long periods of high air mass, effectively negating that advantage. The most reliable way to compare is to run detailed, location-specific energy simulations that incorporate all these factors, rather than relying solely on the nameplate wattage. This holistic approach ensures the selected technology will deliver the maximum possible return on investment over the system’s entire lifetime, accounting for the ever-changing color of the sunlight it receives.