Monitoring chemical degradation of thermally cycled glass-fibre composites using hyperspectral imaging

Abstract

Nowadays, the application of glass-fibre composites in light-weight structures is growing. Although mechanical characterizations of those structures are commonly performed in testing, chemical changes of materials under stresses have not yet been well documented. In the present work coupon tests and Hyperspectral Imaging (HSI) have been used to categorise possible chemical changes of glass-fibre reinforced polymers (GFRP) which are currently used in the aircraft industry. HSI is a hybrid technique that combines spectroscopy with imaging. It is able to detect chemical degradation of surfaces and has already been successfully applied in a wide range of fields including astronomy, remote sensing, cultural heritage and medical sciences. GFRP specimens were exposed to two different thermal loading conditions. One thermal loading condition was a continuous thermal exposure at 120°C for 24h, 48 h and 96h, i.e. ageing at a constant temperature. The other thermal loading condition was thermal cycling with three different numbers of cycles (4000, 8000, 12000) and two temperature ranges (0°C to 120°C and -25°C to 95°C). The effects of both conditions were measured using both HSI and interlaminar shear (ILSS) tests. No significant changes of the physical properties of the thermally cycled GFRP specimens were detected using interlaminar shear strength tests and optical microscopy. However, when using HIS, differences of the surface conditions were detected. The results showed that the different thermal loading conditions could be successfully clustered in different colours, using the HSI linear unmixing technique. Each different thermal loading condition showed a different chemical degradation level on its surface which was indicated using different colours.

Analysis of thermal strains and stresses in heated fibre metal laminates

“…Aging of the epoxy is one of the most likely effects at elevated temperature ranges. [21] The second expected effect is that the local temperature increases and the different thermal expansion coefficients lead to local (mechanical) stress concentrations, especially at the heater elements. These locally high stresses might be equivalent to a mechanical fatigue loading in the vicinity of the heater elements in heated FMLs.…”

Quantitative coating thickness determination using a coefficient-independent hyperspectral scattering model

“…It is used in multiple application fields, including medicine [7][8][9], remote sensing [10], agriculture [11], engineering [12] and cultural heritage [6,13]. Although the range of research fields in which hyperspectral imaging has been applied is vast, the existence of a quantitative hyperspectral imaging approach able to provide thickness information of thin layers could offer new applications beyond the aforementioned fields [14].…”

Snapshot hyperspectral imaging probe with principal component analysis and confidence ellipse for classification

“…HSI provides a wealth of information which can then be used for identification and quantification to reveal important information of the imaged scene, 14 by applying data analysis techniques such as principal component analysis (PCA) and spectral unmixing. 15 HSI has been used to predict beef tenderness, 16 monitor chemical degradation of glass-fiber composites 17 and measure oxygen saturation in the retina. 18 The configurations of HSI systems can be broken down into three main categories, namely the spatial-scanning, spectralscanning and snapshot imagers.…”