Speaker
Description
Ultra-thin recycled polyethylene terephthalate (rPET) polymer films, with thicknesses ranging from 20 to 200 µm, are widely used in applications such as coatings, packaging, surface functionalization, and labelling. Several industrial deposition processes operate at non-ambient temperatures, typically between 20 and 120 °C, making accurate temperature monitoring of these films essential. However, the thermo-optical properties of rPET may vary significantly due to differences in pellet origin, recycling technologies, virgin-to-recycled PET ratios in the formulation, and local thickness variations, which challenges conventional temperature measurement methods.
At such small thicknesses (from 20 to 200 µm), standard infrared thermography is strongly affected by semi-transparency over a broad spectral range (from 0.4 to 20 µm). The thermal flux emanates throughout the entire thickness of the film, which may exhibit an anisothermal temperature distribution. The determination of a precise temperature at a specific physical point within the material’s thickness presents a significant challenge, necessitating the implementation of sophisticated inversion methodologies and multispectral detection techniques that fall beyond the purview of the present study. Secondly, radiation transmitted through the film may contaminate the measured signal, leading to significant temperature errors. These effects become negligible in the infrared only for thicknesses exceeding approximately 1 mm.
This work presents the development of a spectrally selective thermographic method specifically designed for ultra-thin rPET films. The approach relies on the use of an interference filter targeting the strongest infrared absorption bands of rPET, which are associated with the molecular vibrational modes of the polymer. By restricting the measurement to these spectral regions, effective opacity is achieved even for very small thicknesses, thereby restoring the assumptions required for classical infrared thermography.
The study begins with the spectral characterization of rPET using FTIR spectroscopy, enabling the quantitative identification of absorption coefficients (in cm-1) through transmission and reflection measurements. At the selected absorption band, the transparency of rPET diminishes significantly, rendering it effectively opaque when its thickness exceeds 10 µm.
Subsequently, the µ-bolometer technology IR camera FLIR® A655SC, operating in the [7.5-13] µm spectral range, outfitted with the designated interference filter, undergoes calibration across the temperature range [30–150] °C by means of a standardized blackbody reference. The impact of spectral filtering on sensor performance (sensitivity, NEDT, and usable temperature range) is then assessed. Even though the dynamic range of the IR camera exhibits significant degradation, the implementation of an appropriate calibration model effectively minimizes interpolation artifacts, thereby ensuring high fidelity in the resulting data.
Finally, the proposed method is validated experimentally on an ultra-thin rPET film heated by an infrared emitter, with thermographic measurements compared to temperatures obtained from a contact sensor. The observed temperature deviation across the entire thermal range remains consistently below 10°C.
The methodology presented herein demonstrates potential for broader applicability, enabling its extrapolation to alternative polymer film formulations. This approach pave the way for the realization of quantitative temperature-field mapping on semi-transparent polymer films under industrially relevant operational conditions.
