Here we describe the ultrasound-assisted extraction of the phenolic compounds from spruce wood bark and present a straight-forward experimental planning method, allowing the optimisation of the process. The effect of ethanol concentration, temperature and extraction time were evaluated through a 3(2)·2 experimental planning. The efficiency of the extraction process was appreciated based on factorial ANOVA results. The maximum extraction yield of total polyphenols (13.232mg gallic acid equivalents (GAE)/g of spruce bark tested) was obtained using a process time of 60min, an extraction temperature of 54°C and a concentration of ethanol of 70% respectively. These results indicate that an important quantity of bioactive compounds can be extracted from spruce wood bark by ultrasound assisted extraction technology.

In this paper, the kinetics of polyphenols extraction from spruce bark (Picea abies) under ultrasounds action was investigated. Studies were performed in order to express the effect of some specific parameters (as: ultrasounds, surface contact between solvent and solid, extraction time and temperature) on the total phenolic content (TPC). Experiments were performed in the presence and absence of ultrasounds, using different contact surfaces between solvent and solid, for times from 5 to 75min and temperatures of 318, 323 and 333K. All these factors have a positive influence on the process, enhancing the extraction rate by recovering higher amounts of polyphenols. The process takes place in two stages: a fast one in the first 20-30min (first stage), followed by a slow one approaching to an equilibrium concentration after 40min (second stage). In these conditions, the second-order kinetic model was successfully developed for describing the mechanism of ultrasound-assisted extraction of polyphenols from P. abies bark. Based on this model, values of second-order extraction rate constant (k), initial extraction rate (h), saturation concentration (Cs) and activation energy (Ea) could be predicted. Model validation was done by plotting experimental and predicted values of TPC's, revealing a very good correlation between the obtained data (R(2)>0.98).