Premium
Using infrared spectroscopy in combination with multivariate data analysis to predict residual strength of carbon fiber reinforced polymers after one‐sided thermal loading
Author(s) -
Vetter Tanja Marina,
Zimmer Felix,
Eibl Sebastian,
Gudladt HansJoachim
Publication year - 2021
Publication title -
polymer composites
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.577
H-Index - 82
eISSN - 1548-0569
pISSN - 0272-8397
DOI - 10.1002/pc.26136
Subject(s) - delamination (geology) , materials science , composite material , residual , attenuated total reflection , fourier transform infrared spectroscopy , infrared spectroscopy , spectroscopy , carbon fiber reinforced polymer , polymer , fibre reinforced plastic , infrared , optics , composite number , chemistry , paleontology , tectonics , physics , algorithm , organic chemistry , quantum mechanics , computer science , biology , subduction
In this study, structural, mechanical, and chemical changes of one‐sided thermally loaded carbon fiber reinforced polymers (CFRPs) are investigated. The aim is to test and reliably predict residual strength and delamination depth by using infrared spectroscopy. CFRP of different thicknesses (HexPly® 8552/IM7) were irradiated at varying heat fluxes over various time intervals. The inhomogeneously distributed matrix degradation was analyzed by means of attenuated total reflection Fourier transform infrared spectroscopy with a depth resolution of 0.2 mm. Residual interlaminar shear strength (ILSS) was determined and microfocused computed X‐ray tomography was used to measure the delamination depth. Principal component analyses were performed to show which information in the infrared spectra is affected by thermal loading. Furthermore, the combination of spectra taken at different depths of the CFRP can be used to develop partial least squares regression models to predict ILSS and delamination depth. Despite an inhomogenous distribution of thermal damage, precise predictions of ILSS and delamination depth with models considering varying sample thicknesses and heat fluxes were achieved.