In the optical coating industry, performance differences are often measured in fractions of a percent. For anti-reflective (AR) coatings on solar glass, transmission improvements below 0.1% can determine product quality classification. Because the transmission gain between coated and uncoated glass depends directly on coating thickness and refractive index, accurate optical characterisation is essential.
Measuring these properties is challenging. Solar glass typically has a textured surface designed to improve light trapping in photovoltaic modules. The texture scatters light into many directions, which makes conventional transmission measurements unreliable. Integrating spheres are therefore commonly used because they collect both specular and diffuse light. However, even integrating-sphere measurements can introduce systematic errors that become significant when evaluating very small performance differences.
To address this limitation, coating manufacturer Convsun collaborated with Eule27 AG and Avantes to develop an integrated measurement system designed for real production conditions rather than ideal laboratory samples. The system measures transmission and reflection simultaneously and derives coating parameters directly from spectral data using optical modeling.
At the centre of the measurement challenge is a known effect in integrating-sphere photometry: substitution error.
Substitution error in integrating spheres
In a conventional integrating-sphere measurement, a reference spectrum is recorded first with an open port. This condition is treated as 100% transmission. However, the open port allows part of the light to escape from the sphere, reducing internal radiance.
When a glass sample is inserted, even a highly transparent one reflects a small portion of the incident light back into the sphere. This slightly increases the internal radiance compared to the reference condition. Because the detector measures total radiance rather than directional intensity, the sample appears slightly more transmissive than it truly is.
Although the difference is small, it becomes critical when evaluating AR coatings where performance differences are measured in tenths of a percent. Accurate coating characterisation therefore requires correcting this substitution error.
Multi-detector correction approach
The developed measurement system eliminates substitution error by monitoring internal sphere radiance. In addition to transmission and reflection detectors, a third detector observes the sphere wall.
When the sample is introduced, all detectors register signal changes. The ratio between the wall-detector signal during the sample measurement and during the reference measurement provides a wavelength-dependent correction factor. Applying this correction produces transmission and reflection spectra that represent the true optical behavior of the coating.
The setup uses a 180 mm PTFE-coated integrating sphere with a 60 mm measurement port. Illumination combines a halogen light source with a UV LED module to maintain signal intensity below 425 nm. Spectra are recorded using an Avantes AvaSpec-ULS2048XL+ spectrometer covering the 380–1100 nm wavelength range. All detectors acquire data simultaneously, preventing alignment drift and temporal variation between measurements.
Extracting coating parameters
Corrected spectra are analysed using thin-film optical modeling software. From combined transmission and reflection data, the refractive index, extinction coefficient and coating thickness can be determined.
The method works for both single-layer and double-layer AR coatings. Using both transmission and reflection provides sufficient constraints for unambiguous parameter extraction. Reflection-only analysis is also possible using total reflectance measurements, allowing coating characterisation independent of substrate thickness.
Measured spectra show close agreement with theoretical calculations, confirming the validity of the correction approach. Engineers can therefore verify coating performance and identify deviations in the coating process.
Benefits for production control
Accurate measurement enables reliable validation of coating performance. Small transmission and reflection differences can now be quantified and documented. This supports communication of performance data to customers and strengthens product verification.
Internally, precise feedback on coating thickness and refractive index improves process control. Deviations can be detected earlier, allowing adjustments to deposition parameters and improving consistency.
Outlook
Although developed for solar glass, the measurement principle is applicable to other coated glass products requiring precise optical validation. As coating designs become more complex and tolerances tighter, eliminating systematic measurement errors becomes increasingly important in both development and production environments.
By combining integrating-sphere correction, simultaneous spectral acquisition and thin-film modeling, the described approach enables reliable optical characterisation under industrial measurement conditions. Accurate measurement not only verifies coating performance but also supports optimisation of the coating process itself.
