Distribution of Hg during sewage sludge and municipal solid waste Co-pyrolysis: Influence of multiple factors

Abstract

Co-pyrolysis is a promising approach to recover energy from sewage sludge (SS) and municipal solid waste (MSW). Hg emission during this process has serious environmental risks. To reduce the environmental impact, orthogonal experiments on the blending ratio, heating rate, pyrolysis temperature, and residence time were conducted during SS and MSW co-pyrolysis. Variance analysis was used to determine the influence and synergetic effects of these factors. Multivariate nonlinear, neural network, random forest, and support vector machine models were used to simulate the Hg distribution based on four parameters, which were later optimized to optimize the Hg fixing ratio in pyrolysis char. The Hg was mainly distributed in the pyrolysis gas and char. The variance analysis results indicate that the blending ratio is the key factor influencing Hg distribution, and there is little synergetic effect among the four factors. Further experiments showed that a blending ratio of 87.5 SS wt% could enhance Hg fixation in char. The neural network model shows the best simulation performance with a mean relative error of 8.92%. The optimal parameters are a heating rate of 7 °C/min, pyrolysis temperature of 300 °C, and residence time of 10 min, resulting in a Hg fixing ratio of 25.68 wt% in pyrolysis char. The simulated Hg fixation characteristics correlate with the experimental results. This study provides insights into Hg distribution under various conditions during co-pyrolysis of SS and MSW. It is hoped that this work can contribute to the control of Hg during the waste treatment and utilization process.

Keywords: Co-pyrolysis; Mercury; Municipal solid waste; Neural network; Sewage sludge; Variance analysis.