Chemometric characterization of thermally accelerated degradation profiles in gel pen ink using FTIR spectroscopy and CARS-PLSR

In this study, a nondestructive analytical framework for characterizing the thermally accelerated degradation of black gel pen ink was developed using Fourier Transform Infrared (FTIR) spectroscopy. To address the challenges of spectral redundancy and baseline drift, an Abnormal Data Identification (ADI) method and Savitzky-Golay smoothing were applied. The Competitive Adaptive Reweighted Sampling (CARS) algorithm was utilized to isolate 81 key spectral variables associated with solvent volatilization. A Partial Least Squares Regression (PLSR) model based on these optimized variables achieved a Standard Error of Prediction (SEP) of 4.74 days over a 63-day thermally accelerated aging period, outperforming the conventional full-spectrum model. Furthermore, temporal spectral analysis revealed that the degradation process followed first-order exponential decay kinetics (y = A·e−kt + C), providing direct physical-chemical evidence of alcohol and amide volatilization. These results provide preliminary evidence that FTIR spectroscopy coupled with the CARS-PLSR algorithm constitutes a promising chemometric pipeline for the quantitative characterization of ink degradation dynamics under controlled thermally accelerated conditions.