LAUSR

A high-throughput method for the determination of ethanol in vodka and cachaça using 96-well-plate digital images was proposed and validated. The standard method consists of beverage distillation, measuring its density using a pycnometer, and converting it into ethanol content. It is simple, but it is time-consuming and susceptive to error. The proposed method exploits the suppression of phenolphthalein ionization by ethanol in an alkaline solution and the fading of the pink solution was converted into ethanol content. It consists in mixing 1 mL of sample with 0.1 mL of an alkaline phenolphthalein solution. 96-well-plate images were acquired using a desktop scanner. Red, green, and blue (RGB) values from the 96 wells were automatically extracted using ReadPlate (ImageJ plugin). Then, RGB values were exported to a spreadsheet that converted these values into analytical signals and calculated the ethanol content in beverages. The ethanol content of cachaças and vodkas was 33-45% (v/v) and it was also the linear range of the proposed method. The method's precision was evaluated using relative standard deviation (RSD). Five cachaças and three vodkas were analyzed. Each beverage was analyzed six times on the same day (intra-day repeatability) and three consecutive days (inter-day repeatability) by three different analysts (inter-analyst repeatability). The intra-day repeatability average was 1.7% (1.2-2.2% range), the intra-day repeatability average was 2.6% (1.9-3.5% range), and the inter-analyst repeatability average was 4% (2.6-6.2% range). Accuracy was evaluated by comparing the proposed method with the standard method using a percent error and a paired t-test. The average percent error was 1.9%, in the paired t-test, the p-value average value was 0.25. The proposed method can analyze 12 samples in 30 min, whereas the standard method spends around 1 h on each sample. Thus, the proposed method provides high-throughput compared with the standard method.