Optimization of hybrid green extraction techniques for bioactive compounds from citrus lemon peel using response surface methodology (RSM) and artificial neural network (ANN)

Citrus limon peel is a noteworthy byproduct rich in polyphenols, flavonoids, and antioxidants. Conventional processing operations have limitations of long processing time and excess solvent usage. This study explores the utility of pre-treatment with microwave and ultrasound-assisted extraction to enhance bioactive compound extraction. Response Surface Methodology (RSM) and Artificial Neural Network (ANN) were compared for modeling accuracy and predictive performance. Both RSM and ANN modelling showed interactive effects between bioactive component yields and processing parameters. Comparative analysis showed that RSM provided superior prediction accuracy for the given dataset than ANN, although both models performed satisfactorily. The optimized condition including a microwave power of 516.74 W, an irradiation time of 101.86 seconds, ultrasound temperature of 40°C, and ultrasound duration of 21.033 minutes, which yielded a maximum recovery of 2283.72 mg GAE/100 g for TPC, 987.58 mg QE/100 g for TFC, and 78.21% DPPH scavenging activity, with a desirability of 0.99. Additionally, our extraction approach demonstrated a substantial reduction in energy consumption (23.42%) lower than only ultrasound-assisted extraction and CO 2 emissions, which is a marker for environmental sustainability. This study confirms the efficiency of microwave pretreatment and ultrasound-assisted extraction which is green and scalable method for recovering bioactives from Citrus limon peel. • A hybrid green extraction method combining MAE and UAE was applied to enhance bioactive compounds. • RSM showed better predictive accuracy than ANN for extraction modeling. • High bioactive yield was achieved using RSM with 2283.72 mg GAE/100 g TPC, and 987.58 mg QE/100 g TFC. • Energy use was significantly reduced by 23.42% while also lowering CO 2 emissions.