C. M. Henrique, R. F. Teófilo, F. A. L. Ribeiro, M. P. Cereda, M. M. C. Ferreira, "Prediction of Permeability of Modified Cassava Starch Films by FTIR Spectroscopy and PLS". Lisbon, Portugal, 20-23/09/2004: Ninth International Conference on Chemometrics in Analytical Chemistry (CAC IX, CAC-2002). Program and Abstracts Book, (2004). Poster 185. Section: Poster Session 4 - Chemometrics in Food, Environmental & Safety Issues.

ABSTRACT

Prediction of Permeability of Modified Cassava Starch Films by FTIR Spectroscopy and PLS. Celina M. Henrique², Reinaldo F. Teófilo¹, Fabiana A. de L. Ribeiro¹, Marney P. Cereda², Márcia M. C. Ferreira (PQ)*¹ *marcia@iqm.unicamp.br ¹Laboratório de Quimiometria Teórica e Aplicada, Instituto de Química, UNICAMP, CAMPINAS, SP, BRAZIL. ²Faculdade de Ciências Agronômicas de Botucatu, UNESP, BOTUCATU, SP, BRAZIL.  Keywords: cassava starch film, multivariate calibration, PLS.  Biodegradable packaging for food conservation needs to have the same efficacy in the shelf time and in the inhibition of humidity, oxygen, carbon dioxide, aromas, and lipids migration as the conventional packaging (derived from plastic material). The research about this type of product is currently a great challenge to researchers from different areas. The literature on biodegradable films based on native starch is relatively abundant, but is less frequent about modified starches, which really present different properties if compared to regular or native starches.The purpose of this study was to build a multivariate calibration model using infrared spectroscopy (FTIR), with the aim to predict the water vapor permeability of different types of modified films. Casting films were prepared from commercial cassava starch native and modified. The modified starches used were: pre-gelatinized and cross-linking, Carboxy-metyl-starch with low and high viscosity and esterified. The permeability was obtained through the gravimetric method proposed by Gontard (mean value of 0.17 g.mm/m2.h.kPa ± 0.047). The evaluated concentrations in the filmogenic suspension were 3 and 5 % for each starch film and the FTIR spectra were recorded in the region of 4000 to 650 cm-1, with a resolution of 2 cm-1. The variables were mean-centered and selected by the Interval Partial Least Square (IPLS) algorithm. The spectra were pre-processed using the 2nd derivative. The PLS model was validated by cross validation in a set of 34 samples and also by an external set of 9 samples. The best calibration model was obtained with four factors and the PRESS of validation, SEV and r values were equal to 0.015, 0.021 and 0.965, respectively. The selected spectral region was in the range of 3167 to 2701 cm-1 which correspond to the –OH stretching (3500 to 2700 cm-1) and the –Csp3–H stretching (2980 to 2700 cm-1). The first is a polar group and directly proportional to the water vapor permeability while the second is non-polar and inversely related.The estimated permeability values for the external validation set were below 6 %, except for two samples where these values were in the range of 10-12 %. Acknowledgements: The authors thank the agencies CNPq and CAPES, and also Dr. Nathalie Dupuy from (GOAE) - Géochimie Organique Analytique et Environnement, Université de Marseille 3, France, for the FTIR analysis of starchy films.

Keywords: cassava starch film, multivariate calibration, pls

Requested contribution format: poster

Corresponding author: MÁRCIA M. C. FERREIRA (marcia@iqm.unicamp.br)