MODELS OF LEAD MIGRATION OF TWO CELLULOSE CONTAINERS IN CONTACT WITH FOOD SIMULANT

Abstract

The objective of the research was to estimate mathematical models for lead migration (μg/l) from a cellulose matrix using a neutral pH food simulant, taking two independent variables: total migration (mg/dm2 ) and the contact area (dm2 ), using multiple linear regression. The evaluated samples were two cellulose containers destined to the contact with food, being the origin of the cellulose from Chile and Brazil. Total and specific migration tests were carried out with the methodologies of Resolution No. 12 of MERCOSUR (1995) and method 3111 B of APHA (2005) respectively, under conditions of prolonged contact and at a temperature of 50 °C. The results of this research show that only the model for Brazilian pulp (Y = -55,864 – 0,304X1 + 9,173X2) was statistically adequate to make forecasts, with its correlation coefficient (R2) of 0,899. From the descriptive statistical analysis, the variable specific migration of lead with respect to the variables total migration and area presented a linear relationship, according to the Pearson (0,854/0,944) statistic. The model for Chilean pulp (Y = 0,113 + 1,386X1 – 0,085X2) obtained a correlation coefficient (R2) of 0,615 and a moderate linear relationship between the variables according to the Pearson (0,784/0,538) statistic, being the model not suitable for forecasting.