Modeling hydrogen peroxide bleaching to predict optical properties of bleached hardwood CMP

Abstract

Due to the variations in opacity and brightness of peroxide bleached pulp at Mazandaran Wood and Paper Industries Company (MWPI), empirical models were developed to predict chemimechanical pulp (CMP) brightness and opacity from peroxide bleaching conditions and to drive the optimum operating conditions. To overcome the inconsistency problem, a multi-variate regression analysis method was used for model building. The models were then validated using a new data set from the bleach plant at MWPI, assessing the models’ predictive ability and performance. The results show that there is a relationship between bleaching variables and such dependent variables as pulp brightness and opacity. In addition to the hydrogen peroxide charge and pulp initial brightness, the initial opacity had a significant reverse effect on the final CMP brightness. It was also found that the concentration of total Na+ in the CMP tower was the most important variable affecting the final pulp opacity. The validation results demonstrated that these models can be employed as useful tools for process optimization purposes.