Robust and Efficient Anaerobic Digestion of Lignocellulose

Restricted (Penn State Only)
- Author:
- Hirl, Katharine
- Graduate Program:
- Agricultural and Biological Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- February 25, 2022
- Committee Members:
- Jay Regan, Co-Chair of Committee
Juliana Vasco-Correa, Major Field Member
Tom Richard, Co-Chair & Dissertation Advisor
Costas Maranas, Outside Unit & Field Member
Jeffrey Catchmark, Professor in Charge/Director of Graduate Studies - Keywords:
- anaerobic digestion
switchgrass
bioenergy
robustness - Abstract:
- Biofuels derived from lignocellulosic feedstocks have been shown to have zero or net negative life-cycle greenhouse gas emissions while also improving soil and water quality. The positive environmental impact of these fuels has generated interest in producing lignocellulosic biofuels on an industrial scale. Anaerobic digestion (AD) is a biological process that can produce methane from a variety of feedstocks including lignocellulosic biomass. AD uses a mixed microbial community adapted to the feedstock and operating conditions. To improve the viability of AD for lignocellulosic biofuel production, an efficient and robust community is required. While AD systems have been studied for decades, few studies have performed comprehensive analyses across the range of possible operating conditions, nor have there been many controlled studies of the types of perturbations that occur in industrial settings. This research fills this gap by using a combinatorial design of temperature, pH and retention time conditions. The steady state performance of AD communities was evaluated in terms of key process variables, gas production and composition, volatile fatty acid concentrations, total and volatile solids concentration, and overall feedstock conversion. A total of eighteen different conditions from the combination of two temperatures (55°C and 37°C), three pHs (5.5, 7.0, and 8.5), and three retention times (3.3 days, 5 days, and 10 days) were tested. The experimental conditions clustered into four functionally unique groups. Half of the conditions were in a low performing group, the second largest cluster was of medium performing conditions, and the last two groups both were high performing and each contained one condition. The two high performing conditions were both thermophilic and had a ten day retention time, but were distinct in terms of pH and resulted in very different product formation. The alkaline pH 8.5 system favored volatile fatty acid (VFA) production and the neutral pH 7.0 system favored biogas production. The high performing condition favoring VFA production resulted in the highest feedstock conversion at 45.6%. That high performing condition’s percent feedstock conversion was approximately twice that of the medium performing group, and 17% greater than the other high performing condition that favored biogas formation. This highest feedstock converting experimental condition was then subjected to a range of pH and temperature pulse stresses and process variable response was monitored for recovery to assess functional robustness. Across the range of pH (±2.0) and temperature (±30°C) stresses tested, no catastrophic failure was observed indicating that the community as a whole is functionally robust. Thus, this research shows that for lignocellulosic feedstock an AD process operated under alkaline thermophilic conditions can be more beneficial for conversion compared to traditional neutral pH operating conditions. Through a series of pulse stresses of both temperature and pH, this unique set of process conditions was shown to be functionally robust, further increasing its viability for industrial application. Finally, a technoeconomic analysis on a two stage digester system converting manure and switchgrass to RNG was performed using the novel high performing alkaline condition favoring VFA formation as the first stage and the high performing neutral pH condition favoring biogas production as the second stage. This system was compared to a conventional 20 d RT mesophilic single stage system converting the same feed to RNG. The analysis found that for a 1,000 cow dairy farm the two state process is more economically favorable than the conventional system and could become profitable with the addition of low carbon fuel standard credits for carbon offsets.