Melting/crystallization Mechanism of Biodegradable Polymer, Poly (3-hydroxybutyrate), Studied by Quantification of Temperature-dependent IR spectra by Nonlinear Deconvolution

The melting and crystallization mechanisms of poly(3-hydroxybutyrate) (PHB), a biodegradable polymer, were investigated using temperature-dependent infrared (IR) spectroscopy. The IR spectra were quantified by DeNoLeS-PV (Deconvolution by Nonlinear Least Squares with Pseudo-Voigt function) to analyze the thermal behavior of the crystal structure stabilization between PHB helical molecules via hydrogen bondings. The wavenumber region of 4000-1000 cm-1 was utilized. The DeNoLes-PV algorithm played a key role in separately evaluating the absorbance, peak position, and Full Width at Half Maximum (FWHM) for 37 spectra, comprising 29 peaks at 16 different intervals during the temperature increase and 21 intervals during the temperature decrease. The temperature-dependent absorbance profiles of the peaks were classified into four distinct groups using Principal Component Analysis (PCA). The behavior of the amorphous band at 1740 cm-1 and crystalline C = O stretching band at 1723 cm-1 were monitored to investigate the mechanisms of PHB melting and crystallization. The melting process involves the breaking of C-H center dot center dot center dot O = C hydrogen bonds between the molecular chains in lamellar crystals, whereas the cooling process involves a two-stage change due to fluctuations in nucleation. At a temperature of 125 degrees C, a significant change in absorbance appears to be associated with spike-like shifts in the peak position and FWHM. This suggests that the hydrogen bonding structure undergoes a high- dimensional transformation between the crystalline and amorphous regions. The proposed method successfully captured and visualized the melting/crystallization processes of PHB, which involves the decomposition and formation of a higher-order structure comprising a mixture of crystalline and amorphous regions within the lamellar crystalline region. The proposed deconvolution approach can be applied to various instrumental analysis datasets for peak quantification.