Beyond Lolium perenne: the impact of grassland biodiversity on butyric acid and biomethane production in a forage-based anaerobic biorefinery

Abstract

Permanent grassland fields are complex ecosystems consisting of graminoids (e.g., grasses) and forbs (e.g., legumes and herbs), which are crucial for environmental health and food production. Forbs are important players in this ecosystem to the maintenance of grassland’s biodiversity and functional ity. In the context of biorefinery, graminoids and forbs represent a low-cost, abundant and renewable source of feedstock, which can be used for the production of a myriad of products (e.g., biogas, biomethane, volatile fatty acids [VFAs] and fertiliser). In the past 20 years, grasses have been extens ively evaluated as a potential feedstock for anaerobic digestion (AD) due to their sustainability and abundance. However, forbs have been overlooked, despite their importance to the environment and their higher nutritional value compared to grasses. Beyond biogas, the AD process can be tailored to improve the production of selected acids with a higher market value than biogas/biomethane, such as butyric acid and caproic acid. Shifting the process to the accumulation of VFAs would be important to associate resource recovery, sustainability and competitiveness. Few works have studied the use of grassland biomass, fresh or ensiled, to selectively accumulate butyric acid, and none have studied the potential of clovers for VFA production. Moreover, silage quality and its effect in the fermentation of grassland biomass is still unclear despite the importance of feedstock type and quality to VFA production. Therefore, this Ph.D. investigates the use of AD technology in the valorisation of permanent grass lands, evaluating biomethane production and the selective accumulation of butyric acid. This work also aims to elucidate the microbial community dynamics in the fermentation of grass, especially considering different types of silage. The thesis was structured in five chapters. In the first chapter, an introduction to the motivation of this thesis, its research questions and objectives are outlined. In the second chapter, a critical review published in the special issue ‘Anaerobic Fermentation’ – A Biolo gical Route Towards Achieving Net Neutrality (Fermentation, August 2022) is presented in its original form. The review explored the potential for resource recovery from food waste, agricultural and an imal residues and some of the recent advances in the production of VFAs from these substrates. The important notes on chemical parameters, state-of-art systems, and applications were also provided. The main drawback in VFA production, VFA recovery, was also discussed, highlighting a range of challenges and opportunities, which are the current drivers in the field. In the third chapter, a study was designed to evaluate the methane potential of six grassland spe cies adapted to temperate climate: Lolium perenne, Phleum pratense, Trifolium pratense, Trifolium repens, Chicorium intybus, and Plantago lanceolata. The influence of nitrogen fertilisation rates on the methane production was investigated using plots of L. perenne supplied with two levels of nitrogen fertiliser. A biomethane potential assay was performed to elucidate the species identity effects and interspecific interaction effects of mono-digesting those species and co-digestion two or six of those species. Those effects were evaluated using a diversity-interaction model. Synergistic effects were observed in the co-digestion of both T. repens and C. intybus, and L. perenne and T. pratense. Antagonistic effects, though, were observed in the co-digestion of grass species. Improved methane yield was observed in the mono-digestion of L. perenne with a higher dose of nitrogen fertiliser. However, a comparable area-specific methane yield was reached in the co-digestion of an equi-proportional mixture of six grassland species. In the fourth chapter, two fermentation assays were designed to investigate the production of VFAs and butyric acid from L. perenne (two doses of nitrogen fertiliser), T. pratense and T. repens. The co fermentation of L. perenne and T. repens was also evaluated. Leachate and digestate from a pilot-scale leach-bed reactor producing VFAs from grass silage and cattle slurry were used as inoculum. The one step fermentation assay was performed in batch-mode at mesophilic temperatures for 7 and 14 days. In the two-step fermentation assay, the substrate was immersed in water for 6 days before introducing the inoculum to the fermented liquid and fermented solid for further 7 days. The buffering capacity of clovers was an important asset for the biomass degradation, but favoured the accumulation of methane instead of VFAs. Doubling the dose of nitrogen fertiliser did not improve the production of VFAs or butyric acid from L. perenne, but improved the total product yield due to a higher accumulation of methane. No difference in VFA production was observed between the one-stage fermentation and the two-stage fermentation, but results from the two-stage correlated the lactic acid consumption to butyric acid production. The best performance in terms of butyric acid was observed in the one-stage fermentation of L. perenne (low dose of nitrogen fertiliser) or its mixture with T. repens. In the fifth chapter, the microbial community dynamics in the fermentation of grass, and the in fluence of silage pH were investigated using three leach-bed reactors. Each was inoculated with a seed-inoculum based on their potential for silage degradation: rumen fluid, rumen solid and granu lar sludge (reactor 1); cattle slurry and granular sludge (reactor 2); and granular sludge (reactor 3). Silages with pH 4.3, 4.6, 6.7 and 8.1 were tested. The leach-bed reactors were assembled with 5 L working volume and a pH control system, one for each seed-inoculum mixture. The system was operated with a solid retention time of 6 days in sequential batch mode for 8-9 months with constant pH at approximately 5.5. Silage pH was responsible for the shift in VFA profile and yield, with a lower pH (4.3 and 4.6) favouring the accumulation of acids and a high pH (8.1) leading to a very low concentration of VFAs. Higher concentrations of butyric acid were reached by consuming the lactic acid accumulated in the leachate due to the fermentation of the low pH silage. High-throughput sequencing of the 16S rRNA gene in genomic DNA samples and cDNA (synthesised from RNA) samples extracted from digestate and leachate samples reinforced the hypothesis that silage pH was a driving force in the reactor performance. In all reactors, Prevotella and Lactobacillus were the most abundant genera. The higher presence of Lactobacillus associated with the presence of Caproicipro ducens may explain the accumulation of butyric acid and caproic acid by the end of the trial. After the reactor trial, a small batch was performed to investigate the influence of using digestate, leachate or a mixture of both as inoculum in the semi-continuous fermentation of silage. It was concluded that the solid digestate was an important for the degradation of silage, but the combination of leachate and digestate led to a faster VFA production. In summary, this thesis demonstrates the importance of permanent grassland biodiversity and silage preservation in AD, paving the way for potential developments that associate sustainable forage production with VFA production.