Volltext-Downloads (blau) und Frontdoor-Views (grau)
  • search hit 3 of 10
Back to Result List

Characterization of turbid biobased polyurethane thermosets by UV-Vis-NIR spectroscopy combined with multivariate data analysis for in-line process monitoring

  • Polyurethane thermosets have a wide range of applications. In this study, alternative raw materials were used to enhance sustainability. In two newly developed biobased polyurethanes (PUs), the cross-linker content was varied, which caused phase separation and therefore affected the turbidity. To investigate this phenomenon, UV–Vis–NIR spectroscopy was utilized. Spectra were recorded from 200 to 2500 nm in transmittance mode, and multivariate data analysis was applied to the three UV, Vis, and NIR sections separately. For the two different PU classes, each with five different cross-linker contents, classification by principal component analysis combined with linear or quadratic discriminant analysis was possible with an accuracy between 93% and nearly 100%. The best separation was achieved in the NIR range. Partial least-squares regression models were determined to predict the cross-linker content. As mentioned, the model for the NIR range is the most suitable, with the highest R2 (validation) of 0.99 for PU1 and 0.98 for PU2. The corresponding root-mean-square error of prediction values of the external validation was the lowest, with 0.82% (PU1) and 1.25% (PU2). Therefore, UV–Vis–NIR absorbance spectroscopy, especially NIR, is a suitable tool for monitoring the appropriate material composition of turbid PU thermosets in line.

Download full text files

  • 4810.pdf
    eng

Export metadata

Additional Services

Search Google Scholar

Statistics

frontdoor_oas
Metadaten
Author of HS ReutlingenMüller, Sabrina; Knoblich, Mona; Lorenz, Anita; Brecht, Marc; Lorenz, Günter
DOI:https://doi.org/10.1021/acs.iecr.3c01756
ISSN:0888-5885
eISSN:1520-5045
Erschienen in:Industrial & engineering chemistry research : I & EC research
Publisher:American Chemical Society
Place of publication:Washington, DC
Document Type:Journal article
Language:English
Publication year:2023
Volume:62
Issue:40
Page Number:10
First Page:16188
Last Page:16197
PPN:Im Katalog der Hochschule Reutlingen ansehen
DDC classes:540 Chemie
Open access?:Nein
Licence (German):License Logo  In Copyright - Urheberrechtlich geschützt