Predicting Freezing Point of Ice Cream Mix with Infrared Spectroscopy and Chemometrics

Fourier transform infrared spectroscopy (FTIR) has become an increasingly valuable analytical tool in the dairy industry due to its rapid, non-destructive nature and ability to capture complex chemical information. This study evaluated the feasibility of using FTIR in combination with chemometric modeling to predict the freezing point (FP) of ice cream mix (ICM), an important quality attribute that influences product texture, freezing behavior, and manufacturing efficiency. Traditional methods for FP determination are often labor-intensive and can present challenges when analyzing high-solids dairy systems. In this study, FP values were obtained through a cryoscopic method. FTIR spectra were collected for all samples, and partial least squares (PLS) regression models were developed to relate spectral information to FP values. Model performance metrics indicated that FTIR spectra contained information associated with FP variation; however, overall predictive performance was limited. The relatively weak model accuracy was attributed to the complex nature of FP as a colligative property influenced by multiple compositional factors, including sugars, salts, proteins, and stabilizers, as well as the narrow compositional range and limited sample size of the dataset. Despite these limitations, the results demonstrated the potential of FTIR as a rapid screening tool for FP estimation and provided a foundation for future model development using larger and more compositionally diverse datasets and improved reference methodologies.