Multivariate modeling of textile dyes (Reactive Red 195 and Acid Blue 62) photodegradation using a lab-made photoreactor

In this work, we studied the feasibility of Acid Blue 62 (AB-62) and Reactive Red 195 (RR-195) photodegradation employing a lab-made photoreactor. The photoreactor was built up with four low-pressure Hg lamps of 6 W each, and using a medium-density fiber (MDF) box of 13 x 30 x 20 cm in height, length, and width, respectively. The photodegradation process was modeled in a multivariate way by applying a Box-Behnken design. The operational parameters of the process (irradiation time, irradiation power, and hydrogen peroxide concentration) were set as factors for the construction of the models. The factors and the experimental region of the design were selected through a univariate study performed before the application of the Box-Behnken design. According to the multivariate study, both irradiation time and power had a more pronounced effect on the photodegradation efficiency, which was measured through the spectrophotometric determination of the concentration of the dyes in the solutions. The models were built at two concentrations of the dyes, and always presented coefficients of determination (r2) higher than 0.94, indicating an excellent adjustment to the data. The errors of prediction for photodegradation percentage varied from 0.38 to 1.46%. Solutions containing 20 and 80 mu mol L-1 of RR-195 presented a maximum percentage of photodegradation of 98.6% and 85.9%, respectively, whereas AB-62 solutions with 40 and 140 mu mol L-1 achieved percentages of photodegradation of 88.2% and 36.4%, respectively. In all cases, only 4.5 min of irradiation in the presence of H2O2 was necessary to achieve these values.