Abstract :
[en] Energy crops and crop residues can be digested anaerobically either alone or as co-substrate to produce biogas, a versatile renewable energy source. In the agricultural sector, this biomass is often co-digested with animal manure and slurry, one of the most significant agricultural waste streams. In spite of the growing number of agricultural biogas plants using agricultural waste and energy crops, the use of process simulation models for predicting and defining plant behaviour is still very limited, mainly due to the complexity of the process and the heterogeneity of the substrates. Also, the lack of standardized substrate characterisation methodology contributes to this issue. The chemical oxygen demand (COD), widely used as a mass balance term in modelling of wastewater treatment plants, could be applied in this field. However, COD is difficult to measure in the case of heterogeneous and solid substrates, including samples with high solids content. This, in turn, has limited its use to date to evaluate substrate biodegradability over time and methane production potential in the field of co-digestion of agricultural by-products. This research paper reports and analyses experimental results on the applicability of COD as a substrate characterisation and process monitoring parameter in the context of lignocellulosic biomass digestion, as collected from the research work currently being undertaken at the University of Luxembourg. The reliability of this parameter was investigated using different silages. Four sets of batch experiments were carried out using maize silage (12.17gVS/l), grass silage (14.58gVS/l), and two co-digestion mixtures of 40% maize and 60% grass (18.31 gVS/l) and 70% maize and 30% grass (12 gVS/l) in 1 litre reactors running under mesophilic conditions. The reactors were inoculated with pre-acclimatized inoculum from Beckerich agricultural biogas plant in Luxembourg. Soluble and particulate COD, volatile solids (VS), total solids (TS), volatile fatty acids (VFAs), and gas production, including composition, were monitored over a period of 24 days. A few adapted analytical procedures for measuring COD in solid and heterogeneous substrates and samples with high suspended solids concentration, in comparison to standard methods widely used for wastewater, have been proposed in recent years but these have rarely been applied and validated in the field of digestion of agricultural by-products. The COD methodology adopted in this study was developed by the process research group at the University of Luxembourg as a modification of the DIN 38414-S9 considering some suggestions from Raposo et al. (2008). The modified COD method was applied to different substrates (i.e. grass, maize silages, and cellulose) and also to monitor particulate COD (pCOD) evolution during anaerobic digestion of the different silages. The soluble COD (sCOD) method as established in the ISO 15705 standard method was modified for these experiments as the digestate has relatively high suspended solids concentration. First, the measured substrate COD concentration was compared with the theoretical oxygen demand for each substrate. COD recovery obtained was 96% for cellulose, 92.5% on average for the different maize silages tested and 91.5% on average for the grass silages tested. For all different batch series, the pCOD decreased overtime as a result of the hydrolysis of the particulate fraction of the substrate. The pCOD evolution along with the methane accumulative yield suggested that the solubilisation of the substrate was slower in the case of the co-digestion experiments. sCOD, on the other hand, initially increased sharply as a result of the solubilisation of the substrate and subsequent acidification and then decreased as digestion progressed. After 9 days of digestion, 77%, 71%, 69% and 48% substrate to methane conversion rates had been achieved for the experiments fed with grass silage, maize silage, the 70% maize/30% grass mixture and the 40% maize/60% grass mixture, respectively. Measuring the pCOD and sCOD together with biogas and VFAs concentration during digestion allows, not only to estimate the degradation advancement through COD mass balance, but also to obtain relevant information about the solubilisation of the particulate fraction (hydrolysis step).
