首页 | 官方网站   微博 | 高级检索  
     


Classification of cassava starch films by physicochemical properties and water vapor permeability quantification by FTIR and PLS
Authors:Henrique C M  Teófilo R F  Sabino L  Ferreira M M C  Cereda M P
Affiliation:Author Henrique is with Agência Paulista de Tecnologia dos Agronegócios, Piracicaba, SP, Brazil. Authors Teófilo, Sabino, and Ferreira are with Instituto de Química, UNICAMP, Campinas, SP, Brazil. Cereda is with Instituto São Vicente, CeTeAgro/UCDB, Campo Grande, MS, Brazil. Direct inquiries to author Henrique (E-mail: ).
Abstract:ABSTRACT:  Cassava starches are widely used in the production of biodegradable films, but their resistance to humidity migration is very low. In this work, commercial cassava starch films were studied and classified according to their physicochemical properties. A nondestructive method for water vapor permeability determination, which combines with infrared spectroscopy and multivariate calibration, is also presented. The following commercial cassava starches were studied: pregelatinized (amidomax 3550), carboxymethylated starch (CMA) of low and high viscosities, and esterified starches. To make the films, 2 different starch concentrations were evaluated, consisting of water suspensions with 3% and 5% starch. The filmogenic solutions were dried and characterized for their thickness, grammage, water vapor permeability, water activity, tensile strength (deformation force), water solubility, and puncture strength (deformation). The minimum thicknesses were 0.5 to 0.6 mm in pregelatinized starch films. The results were treated by means of the following chemometric methods: principal component analysis (PCA) and partial least squares (PLS) regression. PCA analysis on the physicochemical properties of the films showed that the differences in concentration of the dried material (3% and 5% starch) and also in the type of starch modification were mainly related to the following properties: permeability, solubility, and thickness. IR spectra collected in the region of 4000 to 600 cm−1 were used to build a PLS model with good predictive power for water vapor permeability determination, with mean relative errors of 10.0% for cross-validation and 7.8% for the prediction set.
Keywords:chemometrics    modified starch    PCA    physicochemical properties    PLS
本文献已被 PubMed 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司    京ICP备09084417号-23

京公网安备 11010802026262号