Improvement of co-digesting swine manure and waste kitchen oil
No Thumbnail Available
Authors
Meeting name
Sponsors
Date
Journal Title
Format
Thesis
Subject
Abstract
Anaerobic co-digestion of swine manure (SM) and waste kitchen oil (WKO) was conducted to evaluate the effect of substrate loading rates on biogas production efficiency. Biogas yields of M4O2 (4 g-VSSM L-1 d-1 + 2 g-VSWKO L-1 d-1) and M2O1 were 917 [plus or minus] 43 and 909 [plus or minus] 37 mL g-VS-fed-1, which were 25.7 percent and 24.6 percent higher than the mono-digestion of M2, respectively. However, higher OLRs of SM and WKO did not increase biogas yield. A significant increase of bacterial alpha-diversity was observed in M2O1, at 233.0 [plus or minus] 3.6 compared with 218.7 [plus or minus] 5.1 of M2. Less bacterial alpha-diversity and volatile fatty acids accumulation were observed in M4O1 and M4O2. When the digesters were fed with M2, introduction of more than 1.2 g-VSWKO L-1 d-1 did not increase biogas yield per VS-fed compared to M2O1, and might cause system imbalance. It took up to 84 days to observe system failure. Biogas potential (BP) study showed significantly higher of biogas yield produced from WKO compared to SM. Therefore, addition of WKO into SM digesters resulted in the increase of biogas production, from 9.0 [plus or minus] 3.2 percent to 22.6 [plus or minus] 3.1 percent, compared with digestion of only SM. However, BP of M4O1 and M4O2 were 93.2 [plus or minus] 6.6 percent and 95.6 [plus or minus] 4.8 percent, compared with BP of separate substrates, which indicated that the synergistic effects of co-digestion were due to the increase of organic loading rate and nutrient balance during the process. Co-digesting SM and WKO at ambient or thermophilic working temperature resulted in the low biogas productions compared to mesophilic condition. Application of M2O1 in the on-farm ADs could result in the highest biogas production while maintaining system stability. However, water consumption and digester size were both higher than when M4O1 or M4O2 was applied. A polynomial regression model with variable interaction was developed which showed the effectiveness in terms of biogas production from SM and WKO. The differences of biogas yield estimated by the model and the lab results were in the range of 0.2 percent to 8.6 percent, except only one OLR, in which 15.9 percent difference was observed. However, only the selected dataset with removal of zero biogas values was utilized for the model improvement, and model application was limited in the ranges of each key variable. Two decision support tools were developed based on the model to estimate biogas production and other operating factors of on-farm ADs.
Table of Contents
DOI
PubMed ID
Degree
Ph. D.