Storage of wafer cookies: Assessment by destructive techniques, and non-destructive spectral detection methods

Abstract

Wafer cookies combine two or more layers of wafer sheets with intermediate layers of cream filling and later coating with chocolate. During storage, wafer cookie quality, especially in terms of mechanical properties, is mainly affected by moisture migration from the cream or chocolate and moisture absorption from air. This study aimed to assess the storage of wafer cookies by destructive (water activity, mechanical properties, and sensory acceptance) and non-destructive methods (image analysis, NIR spectroscopy and hyperspectral imaging HSI). Furthermore, two packaging types were considered. Samples were stored at 18 °C (RH = 50%) and analysed after 2, 4, 5, 6, 7 and 8 months. Good linear relations (R 2 up to 0.84) were found between water activity and mechanical parameters, confirming the dependence between textural aspects and water content. By adding a multi-material packaging layer, the shelf life significantly increased in terms of sensory acceptance (crispness). No significant differences were found between the surface colour parameter (white index) attributable to fat bloom formation. PCA results of NIR and HSI spectra showed a clear separation between samples acquired at time 0 and those analysed during storage that was related with the packaging type and storage time. PLS models developed to estimate the storage time showed R 2 ranging from 0.926 (RMSECV = 0.63 months) to 0.960 (RMSECV = 0.52 months), while the water activity ranged from 0.858 to 0.928 (RMSECV = 0.02 months). The PLS models based on HSI spectra were used to obtain predictive images of water activity or storage time. • Storage of wafer cookies was evaluated considering two packaging types. • Water activity, sensory and mechanical parameters were considered. • NIR spectroscopy and Vis/NIR hyperspectral imaging were used as non-destructive techniques. • PCA showed separation between samples at different storage time and packaging type. • PLS models predicted storage time and water activity with R 2 from 0.858 to 0.960