Optimization of food waste hydrothermal liquefaction by a two-step process in association with a double analysis

Bio-oil production from food waste, consisting of pineapple peel, banana peel, and watermelon peel, is investigated by a two-step process, namely, an alkaline pretreatment process with K2CO3 (10 wt% of the dry feedstock) followed by a hydrothermal liquefaction (HTL) process. Meanwhile, the Taguchi method is introduced to maximize the energy yield of the two-step process. Four parameters in the Taguchi approach are taken into account; they are the pretreatment temperature and time as well as the liquefaction temperature and holding time. The optimal combination of the four parameters gives the highest energy yield of 56.55%. The higher heating value of the bio-oil is 25.12 MJ/kg, yielding a 45.88% improvement when compared to the HHV of the dry-basis feedstock. A double analysis, namely, the Taguchi approach and analysis of variance (ANOVA), suggests that the liquefaction temperature plays the most influential role in the energy yield, and a strong linear relationship (R2 ≈ 0.99) is exhibited between the effect in the Taguchi approach and the F value in ANOVA. The experiments of thermogravimetric analysis coupled with Fourier-transform infrared spectroscopy indicate that the composition of the bio-oil from the optimal operation is more uniform.