Synergistic ultrasound-assisted organosolv pretreatment of oil palm empty fruit bunches for enhanced enzymatic saccharification: An optimization study using artificial neural networks

• Published in: Biomass & bioenergy 2020-08, Vol.139, p.105621, Article 105621
• Main Authors: Lee, Kiat Moon, Zanil, Mohd Fauzi, Chan, Kok Keong, Chin, Zhi Ping, Liu, Yee Chian, Lim, Steven
• Format: Article
• Language: English
• Subjects: Artificial neural networks; Delignification; Empty fruit bunches; Organosolv pretreatment; Process optimization; Ultrasound irradiation
• ISSN: 0961-9534; 1873-2909
• DOI: 10.1016/j.biombioe.2020.105621

The efficiency of ultrasonic-assisted organosolv in pretreating oil palm empty fruit bunches (EFB) was investigated in this study. The effect of temperature, time, sonication power, ethanol concentration and presence of different types of catalysts were examined. The ultrasonic-assisted organosolv pretreatment was found to be significantly affected by temperature, time and sonication power. These parameters were further subjected to process optimization study using Leverburgh Marquee artificial neural networks (ANNs). An empirical model with good predictive accuracies (R-squared value of 0.9084) was generated. From the morphology study, pretreated EFB showed significant structural disruption through the breakage of hemicellulose and lignin bonds, leading to the enhancement of enzymatic saccharification. A maximum reducing sugars of 356 mg/g biomass (7.12 g/L) was obtained at optimized conditions of 48.2 °C, 30 min and 55% (192.5 W) sonication power. The relatively high yield of reducing sugars associated with lower lignin content compared to raw EFB also suggested the effectiveness of ultrasonic-assisted organosolv in pretreating EFB. Therefore, this synergistic approach could pave the way for a more efficient and cost-effective pretreatment process for the production of various bioproducts and biofuels in the future. [Display omitted] •The ultrasonic-assisted organosolv pretreatment is effective in pretreating EFB.•The pretreatment was significantly affected by temperature, time and sonication power.•The empirical models developed from Leverburgh Marquee ANN are of good predictive accuracy.•Reducing sugars yield of 356 mg/g biomass was achieved under optimized conditions.