Dataset and multivariate statistical tools for mathematical modeling of water desorption isotherms and infrared spectral properties of cupuassu (Theobroma grandiflorum L.) pulpMendeley Data

Abstract

This study presents a comprehensive dataset of water desorption isotherms and infrared spectral data for cupuassu pulp, a by-product with significant potential for value-added applications in the food industry. The dataset includes experimentally determined desorption isotherms within a water activity range of 0.1–1 at 25 °C, obtained using the Dynamic Dewpoint Isotherm (DDI) method. Infrared spectra were acquired via Attenuated Total Reflectance Fourier Transform Infrared (ATR-FTIR) spectroscopy, covering the wavenumber range of 4000–650 cm–1 with a resolution of 8 cm–1. Additionally, mathematical modeling and multivariate statistical tools were applied to analyze water desorption behavior, optimizing low-temperature energy consumption and ensuring maximum storage stability. Preprocessing techniques, including baseline correction, Standard Normal Variate (SNV), and Multiplicative Scatter Correction (MSC), were employed to enhance spectral data quality. Principal Component Analysis (PCA), based on scores-loadings computation, was utilized for exploratory analysis of the cupuassu pulp spectral fingerprint. The dataset is provided in Excel format, organized by experimental conditions and replicates. Moreover, MATLAB® R2023a (The MathWorks Inc., Natick, MA, USA) scripts for multivariate statistical analysis were implemented to facilitate model-based assessment of water desorption and infrared properties. In this regard, two MATLAB scripts detail the step-by-step mathematical modeling of isotherms using Peleg, Caurie, and White & Eiring sorption models, as well as Gibbs free energy computation for moisture stability optimization. Furthermore, two MATLAB scripts are dedicated to spectral preprocessing and PCA: one prompts the user to select preprocessing techniques, including the option for sequential application, and visualizes the resulting infrared spectra; the other enables PCA calibration based on the selected preprocessed data. This research provides valuable insights for the food industry, supporting informed decision-making in cupuassu pulp processing. By improving food-processing strategies, ensuring product consistency, and optimizing dehydration processes, this dataset contributes to the development of value-added products from cupuassu by-products, promoting sustainability and resource efficiency.

Keywords

• Model-based investigation
• Gibbs free energy
• Low-temperature drying
• Energy requirements
• Non-spontaneous process
• Infrared spectral analysis