Presently the work deals with outreaching and exhaustive investigations involving single factor optimisation method, to optimise the parameters for phenolic antioxidant extraction from the seeds of Tamarindus indica. At first the characterization of the seeds has been performed by estimating the total calorie value, carbohydrate, protein, fat, Total Polyphenol Content (TPC), moisture (%), total ash (%), total solid (%), volatile solid (%), fixed carbon (%), bulk density (%), pH and solubility. The effects of solid to solvent ratio (1:10 -1:30 w/v), ethanol concentration (25 -100% v/v), extraction time (1 -24 hours) and extraction temperature (25 -60°C) have been investigated to optimise the extraction of Total Polyphenol Content (TPC) and Antioxidant capacity determined by Folin -Ciocalteu and FRAP analyses respectively. The solvent extraction conditions have been optimised at solid to solvent ratio of 1:20 w/v, 50% ethanol as solvent and 3 hours of shaking at 40°C.

The seeds of Tamarindus indica are known to possess a wide range of phenolic compounds with high antioxidant activity as measured by the ferric reducing antioxidant power (FRAP). In the present study, the optimum conditions for the extraction of crude phenolic antioxidants from Tamarind seed were determined using response surface methodology (RSM). A central composite design (CCD) was used to investigate the effects of four independent variables, namely concentration of extractable solids in solvent (g/ml; X1), extraction time (h; X2), extraction temperature (°C; X3) and solvent concentration (%, v/v; X4) on the responses of total polyphenol content (TPC) and FRAP. The CCD consisted of 30 experimental runs. A second-order polynomial model was used for predicting the responses. Canonical analysis of the surface responses revealed that the predicted optimal conditions for the maximal yield of TPC and FRAP were concentration of extractable solids in solvent of 0.049 g/ml, extraction time of 3.24 h, extraction temperature of 45 °C and a solvent concentration of 50%. The experimental values in the optimised condition coincided with the predicted ones within a 95% confidence interval, hence indicating the suitability of the model and the success of RSM in optimizing the extraction parameters.