Optimization of Input Variables in Ultrasonic Vibration-Assisted Pelleting of Cellulosic Biomass Using Multiple Response Surface Methodology

Abstract

Cellulosic ethanol is an attractive alternative to petroleum-based liquid transportation fuels. However, low density of cellulosic biomass (the feedstock for cellulosic ethanol) causes high costs in biomass logistics and hinders large-scale and cost-effective manufacturing of cellulosic ethanol. Ultrasonic vibration-assisted (UV-A) pelleting can significantly increase the density of cellulosic biomass by compressing raw cellulosic biomass into pellets. Pellet density and durability are two important physical properties of a pellet. In this study, a multiple response surface methodology was employed to optimize the input variables (pelleting time, pressure, and ultrasonic power) in UV-A pelleting of sorghum stalks for simultaneously maximized pellet density and durability. Second-order polynomial models were used to fit the experimental results. Main and interaction effects of the input variables on pellet density and durability were also investigated.