The aim of this study was to prepare tablets containing ground fruits of cornelian cherry (Cornus mas L.) with high antioxidant capacity. The experiment was planned and evaluated on Design of Experiment (DoE) principle using Multivariate Data Analysis (MVA) as modern tools used in Quality by Design (QbD) approach. Various tableting mixtures with three different particle sizes of the plant material (up to 800 μm, more than 800 μm and their mixture) and percentage of silicon dioxide (1, 3 and 5%) were prepared. Tablets with a diameter of 10 mm and mass of 400 mg were subsequently produced from these mixtures using two compression forces (C1=7 kN and C2=14 kN). Principal Component Analysis (PCA) and Multiple Linear Regression (MLR) with response surface methodology were used to find the influential process-formulation parameters and describe their optimal settings. Finally, it is possible to say that the increasing level of silicon dioxide and the decreasing particle size of ground cornelian cherry lead to prolongation of disintegration time and increase of radial hardness and abrasion loss. Maximal antioxidant activity was obtained using 5% amount of silicon dioxide, the largest particle size and the low compression force.
Keywords: Cornus mas; direct compression; multivariate data analysis; plant extract; response surface.