Artificial Neural Network and Non-Linear Regression-Based Optimization of the Degradation of Acid Black 52 Using Atmospheric Plasma Process.

While plasma technology shows promise for wastewater treatment, its large-scale application is hindered by the challenge of defining universal treatment conditions for complex industrial effluents. This research investigates the degradation of Acid Black 52 (AB52) to overcome this barrier. We treated 250 mL of a 0.5 mM AB52 solution with atmospheric plasma, systematically varying the pH, applied voltage, and treatment time. The process was monitored by using UV-vis absorption spectroscopy, pH, and electrical conductivity measurements. Our findings reveal that plasma treatment can achieve over 80% dye removal under conditions optimized by both the Response Surface Methodology (voltage: 2.57 kV, discharge exposure time: 107.67 min, initial pH: 4.4) and a Genetic Algorithm (voltage: 2.6 kV, discharge exposure time: 107.84 min, initial pH: 4.95). Furthermore, statistical analysis confirmed that the applied voltage and treatment duration are the most influential factors. This work highlights the efficacy of atmospheric plasma and provides a framework for optimizing its use against specific organic contaminants.