Formulation of active food packaging by design: Linking composition of the film-forming solution to properties of the chitosan-based film by response surface methodology (RSM) modelling.

An active chitosan-based film, blended with the hydrolysable tannin-rich extract obtained from fibrous chestnut wood (Castanea sativa Mill.), underwent a simultaneous engineering optimization in terms of measured moisture content (MC), tensile strength (TS), elongation at break (EB), and total phenolic content (TPC). The optimal product formulation for a homogeneous film-forming solution was sought by designing an empirical Box-Behnken model simulation, based on three independent variables: the concentrations of chitosan (1.5-2.0% (w/v)), extracted powder-form chestnut extract (0.5-1.0% (w/v)) and plasticizer glycerol (30.0-90.0% (w/w); determined per mass of polysaccharide). Obtained linear (MC), quadratic (TS or EB), and two-factor interaction (TPC) sets were found to be significant (p < 0.05), to fit well with characteristic experimental data (0.969 < R2 < 0.992), and could be considered predictive. Although all system parameters were influential, the level of polyol played a vital continuous role in defining EB, MC, and TS, while the variation of the chestnut extract caused an expected connected change in affecting TPC. The component relationship formula of chemical mixture fractions (1.93% (w/v) of chitosan, 0.97% (w/v) chestnut extract and 30.0% (w/w) of glycerol) yielded the final applicable material of adequate physico-mechanical properties (MC = 17.0%, TS = 16.7 MPa, EB = 10.4%, and TPC = 19.4 mgGAE gfilm-1). Further statistical validation of the concept revealed a sufficient specific accuracy with the computed maximal absolute residual error up to 22.2%. Herein-proposed design methodology can thus be translated to smart packaging fabrication generally.