Spectral Response Measurement of Photovoltaic (PV) Modules: Principles, Standards and Test Methods

Spectral response measurement determines how efficiently a photovoltaic module converts incident light into electrical current as a function of wavelength, providing critical data for performance calibration, spectral mismatch correction, and energy yield prediction. Performed in accredited third-party laboratories in accordance with IEC 60904-8 and ASTM E1021, the test is applicable to crystalline silicon, thin-film, multi-junction, and emerging PV technologies. This article explains the measurement principles, sample requirements, test items, procedures, and typical applications of spectral response (SR/EQE) testing for PV modules.

Technical Overview

Spectral response measurement, also referred to as spectral responsivity measurement, quantifies the electrical current generated by a photovoltaic device per unit of incident monochromatic optical power at each wavelength. The absolute spectral responsivity s(λ), expressed in A/W, is defined as the short-circuit photocurrent produced by the device divided by the monochromatic radiant power incident on its active area. From this quantity, the external quantum efficiency (EQE) can be derived using the relationship EQE(λ) = 1240 × s(λ) / λ, where s is in A/W and the wavelength λ is in nanometers. The internal quantum efficiency (IQE) can further be calculated by correcting the EQE for the measured reflectance of the device.

The spectral response curve is a fundamental fingerprint of a PV device, since it reveals how effectively the absorber material, junction structure, anti-reflection coating, and encapsulation convert photons across the usable solar spectrum, typically from approximately 300 nm to 1200 nm for silicon-based products and beyond for other material systems. Because the standardized test spectrum (AM1.5G, defined in IEC 60904-3) differs from the spectral output of solar simulators and from real-world sunlight, spectral responsivity data are indispensable for computing spectral mismatch corrections and for accurate energy rating under varying irradiance conditions.

The principal international standards governing this measurement include IEC 60904-8 (Measurement of spectral responsivity of a photovoltaic device), ASTM E1021 (Standard Test Method for Spectral Responsivity Measurements of Photovoltaic Devices), and related standards IEC 60904-7 (computation of the spectral mismatch correction), IEC 60904-10 (linearity measurement), and IEC 61853-2 (energy rating, which requires spectral responsivity as an input parameter).

Testing Samples

Crystalline Silicon PV Modules

Monocrystalline and multicrystalline silicon modules, including PERC, TOPCon, heterojunction (HJT), and bifacial designs, are the most frequently tested sample category. Typical measurement is performed over the 300–1200 nm range, where the spectral response is dominated by front-surface recombination in the short-wavelength (blue) region and by bulk diffusion length and rear-surface recombination in the long-wavelength (near-infrared) region. Encapsulated modules may show reduced blue response due to UV absorption in the glass and encapsulant, which the measurement accurately captures.

Thin-Film PV Modules

Amorphous silicon (a-Si), cadmium telluride (CdTe), copper indium gallium selenide (CIGS), and related thin-film products exhibit pronounced metastable behavior, photoconductivity effects, and capacitive characteristics. Their spectral response is narrower than that of silicon, requiring measurement ranges matched to the absorber bandgap. These devices generally require measurement under white-light bias and controlled bias voltage to reproduce realistic operating conditions, as specified in IEC 60904-8.

Multi-Junction and Emerging PV Devices

Multi-junction III-V concentrator cells, perovskite devices, perovskite-silicon tandems, and organic PV require specialized procedures, including spectrally selective bias illumination to control the current-limiting subcell. Measurements are extended into the near-infrared up to approximately 1400 nm or higher depending on the bottom-cell bandgap.

Cells, Mini-Modules and Reference Devices

Individual cells, laminated mini-modules, glass-glass coupons, and calibrated reference devices are also accepted as test specimens. Reference cells with traceable spectral responsivity certificates, typically calibrated by a national metrology institute, serve as comparison standards during the measurement.

Testing Items

The standard deliverables of a spectral response measurement campaign include:

• Absolute spectral responsivity s(λ) in A/W versus wavelength
• Relative spectral response curve (normalized shape)
• External quantum efficiency (EQE) spectrum in percent
• Internal quantum efficiency (IQE) spectrum, when hemispherical reflectance is also measured
• Short-wavelength and long-wavelength cut-off characteristics and effective bandgap estimation
• Integrated photocurrent density under the AM1.5G reference spectrum, used for cross-checking I-V measurement results
• Linearity verification with respect to irradiance level, where required by IEC 60904-10

Testing Methods

Monochromatic Illumination System

The test specimen is illuminated by a tunable monochromatic light source, typically a xenon or halogen lamp coupled to a grating monochromator, a filter-wheel system, or a scanning array of narrow-band LEDs. The monochromatic beam is modulated by an optical chopper at a frequency typically in the range of 30–200 Hz, and the resulting modulated photocurrent is detected with a lock-in amplifier to reject ambient and stray light contributions. The beam spot, typically a few millimeters in diameter, is directed onto the active area of the device, and its power is monitored continuously with a calibrated photodetector or measured before and after each scan.

Reference Device Comparison Method

The measurement follows the substitution principle defined in IEC 60904-8: the calibrated reference device is placed at the same position as the device under test, and the ratio of the device's modulated short-circuit current to that of the reference device yields the absolute spectral responsivity, eliminating the need for absolute optical power metrology at every wavelength. Reference devices are calibrated at national metrology institutes such as NIST, PTB, NIM, AIST, or NPL, ensuring metrological traceability of the results.

Bias Light and Bias Voltage Conditions

For devices exhibiting non-linear behavior, including most thin-film technologies, the measurement is performed under simultaneous broadband white-light bias superimposed on the modulated monochromatic probe beam. The bias illumination replicates one-sun operating conditions so that the differential responsivity measured with the lock-in technique corresponds to the true operating spectral response. An applied bias voltage, commonly maintaining the device at short-circuit conditions despite its internal series resistance, is frequently required for thin-film and large-area modules, which exhibit significant capacitive or distributed-resistance effects.

Test Procedure

• Stabilize the specimen at 25 ± 1 °C using a temperature-controlled stage or enclosure, and, for thin-film samples, complete any prescribed light-soaking pre-conditioning.
• Verify system performance by measuring a calibrated reference device and confirming agreement with its certificate within the stated uncertainty.
• Mount the device under test, align the monochromatic spot on the active area, and apply any required bias light and bias voltage.
• Scan the wavelength range (for example 300–1200 nm for silicon) in steps of 5 nm or 10 nm, recording the modulated photocurrent, monitor-detector signal, and device temperature at each point.
• Repeat scans at multiple locations on the device to assess spatial uniformity where required, and repeat the reference measurement to verify system stability.

Data Processing

The raw current ratios are converted to absolute spectral responsivity, corrected for monitor-detector drift, spot-size differences, and filter-bandwidth effects. The spectral responsivity curve is then used to compute the EQE spectrum, and numerical integration of the EQE with the AM1.5G reference spectral irradiance yields the expected short-circuit current density. For calibration purposes, the same data are used to calculate the spectral mismatch factor according to IEC 60904-7. Reported measurement uncertainties are typically within ±2 % to ±5 % (k = 2) over most of the spectral range, with larger uncertainties near the UV and long-wavelength cut-off regions.

Applications

Spectral response data support a wide range of engineering, quality assurance, and research activities across the photovoltaic industry:

• Spectral mismatch correction of solar simulator measurements and calibration of reference modules and cells
• Input data for PV module energy rating and annual energy yield simulation according to the IEC 61853 series, accounting for site-specific solar spectra
• Design qualification and type approval testing supporting IEC 61215 and IEC 61730 sequences for crystalline and thin-film modules
• Technology comparison and incoming inspection for utility-scale projects, matching module spectral characteristics to local spectral conditions such as high-altitude or high-moisture sites
• Failure and degradation diagnostics: blue-response loss indicates surface or emitter damage and UV-induced encapsulant degradation, while red-response loss signals bulk lifetime degradation, cracks, or rear-side passivation failure
• Evaluation of light-induced degradation (LID), light- and elevated-temperature-induced degradation (LeTID), and potential-induced degradation (PID) through before-and-after spectral comparison
• Optimization of anti-reflection coatings, cell passivation layers, and tandem cell current matching in research and development programs
• Characterization of bifacial modules, indoor photovoltaic products, and devices operating under artificial or filtered light spectra

FAQ

What is the difference between spectral response and quantum efficiency?

Spectral response expresses the generated photocurrent per incident optical power in A/W, while quantum efficiency expresses the fraction of incident photons converted into collected electrons. The two quantities are mathematically equivalent and interconvertible through the wavelength, with EQE(λ) = 1240 × s(λ) / λ when λ is in nanometers and s is in A/W.

Which standards govern the measurement?

The primary standard is IEC 60904-8, which specifies requirements for the light source, reference device, bias conditions, and data analysis for both linear and non-linear devices. ASTM E1021 is the equivalent method commonly used in North America. Supporting standards include IEC 60904-7 for mismatch computation, IEC 60904-10 for linearity, and IEC 60904-3 for the reference spectral irradiance distribution.

Why do thin-film modules require bias light and bias voltage?

Thin-film devices such as a-Si, CdTe, and CIGS exhibit superlinear or sublinear photoconductivity, metastable states, and internal capacitance. Without white-light bias, the differential responsivity measured by the lock-in technique would not represent real operating conditions. A bias voltage is applied to compensate for series-resistance effects and hold the device at an effective short-circuit condition across the scan.

What wavelength range and step size are used?

For crystalline silicon modules, the typical range is 300–1200 nm with 5 nm or 10 nm steps. Amorphous silicon and CdTe devices are usually scanned from 300 nm to their absorption edge near 800–900 nm, while CIGS, multi-junction, and tandem devices may require scanning up to 1300–1400 nm or beyond.

Can a full-size module be measured, or only cells?

Both are possible. Full-module spectral response systems illuminate a representative cell region or the entire module with large-area monochromatic beams, while the surrounding cells are held under uniform bias illumination so that the measured response reflects true module-level behavior. Individual cells and mini-modules are measured with standard small-spot systems.

What sample preparation and turnaround are typical?

Specimens should be electrically intact with accessible terminals and clean front surfaces. Thin-film samples may require controlled light soaking before measurement. A complete spectral response characterization, including setup, stabilization, scanning, and reporting, is generally completed within one to three working days per specimen in an accredited third-party laboratory.