Process disturbances in anaerobic digestion : biogas-based detection and chemically induced recovery strategies assessed through metabolic and microbial monitoring
(2025)
Files
Thesis_Sébastien_Lemaigre.pdf
Open Access - Adobe PDF
- 11.16 MB
Details
- Authors
- Supervisors
- Abstract
- Agricultural biogas production via anaerobic digestion (AD) of organic feedstocks offers strong potential to mitigate climate change while recycling essential nutrients. However, the AD process is vulnerable to biological disturbances that can cause major economic losses for biogas plant operators. The two main disturbances are free volatile fatty acid (FVFA) intoxication from organic overload and free ammonia nitrogen (FAN) intoxication from nitrogen overload. In full-scale plants, these often disrupt biogas production due to the absence of robust, low-cost online monitoring. Additionally, bioaugmentation methods for restoring critically intoxicated AD reactors remain difficult to scale from laboratory to industry. This thesis investigated biogas-based online monitoring for disturbance detection and scalable recovery strategies for critically intoxicated reactors. In two successive pilot-scale experiments combining metabolic and microbial monitoring, three reactors were progressively subjected to organic or nitrogen overload, leading to FVFA or FAN intoxication, while a fourth reactor was maintained in steady state as a reference. Once methane production ceased in the intoxicated reactors, chemical recovery methods were developed and tested. Post-experiment, principal component analysis-based multivariate statistical process control (PCA-MSPC) models using biogas composition were built and evaluated. Biogas-based PCA-MSPC provided valuable early warning signals when trained with data from an independent, steady-state reactor. Static models, based mainly on biogas composition, enabled early detection of FVFA intoxication, while recursively updated models, using biogas composition alone, allowed early detection of FAN intoxication. In the latter case, warning signals coincided with major shifts in the reactor microbiome, indicating that biogas composition reflects microbial stress responses. Recovery from critical FVFA intoxication was achieved without re-inoculation, simply by neutralising pH with NaOH and restoring buffer capacity with NaHCO₃. FAN intoxication was overcome through a multi-step approach: lowering pH with acetic acid, dilution, and re-inoculation with effluent from a low-nitrogen input reactor. Process stability was provided with non-acclimatised inoculum, avoiding the need for an additional reactor to produce ammonia-acclimatised microbial cultures. In both cases, Methanosarcina spp. likely drove recovery due to their rapid growth, robustness, and metabolic versatility. Our biogas-based monitoring approach, though requiring real-time validation, offers a low-cost means of disturbance detection. Combined with the proposed recovery strategies, it could help biogas plants operate closer to maximum capacity while reducing vulnerability. Future work should assess post-digesters as reference reactors for monitoring and potassium-based chemicals, less harmful to soils than sodium-based compounds, for recovery.
- Affiliations
Citations
Lemaigre, S. (2025). Process disturbances in anaerobic digestion : biogas-based detection and chemically induced recovery strategies assessed through metabolic and microbial monitoring.