The first full-scale Kaumera extraction plants are in operation and increases the circularity of WWTP already. However, to reach a goal of zero waste production it is necessary to look into the waste stream of the Kaumera extraction itself. Roughly 30% of organics is extracted in the process and the remaining organics in the waste can be further recovered using anaerobic digestion (AD). In this study the continuation of the alkaline AD was used, instead of neutral digestion. Main reason for the alkaline digestion compared to the neutral digestion is the increase in CH4 content in the biogas as CO2 remains in the liquid at pH 9.6. The batch incubation uses the alkaline waste residuals stream of the Kaumera extraction plant in Epe. The digestion was done at haloalkaline conditions (pH 9.6; 0.6 M Na+). Combinations of inoculum enriched for similar substrates from a previous study and fresh soda lake sediment were used for the AD batch incubations. CH4 yields varied from 8-28% of total COD going into CH4. Compared to literature this is on lower side as these conversions are in the range of 35-50%. However, some were incubations with pre-treated substrate and already enriched incubations. Others
were neutral digestion of similar substrate and the substrate used in this study. This does show the potential still for a higher conversion of methane in the alkaline digestion of the Kaumera residuals. Based on a titration, the Alkalinity need to keep the pH at 9.6 is 3 g/L of NaOH to prevent a drop from pH 9.64 to 9.34. The overall process observed takes longer than the neutral digestion due to a delay seen in acetate conversion, therefore no bottle-neck in the process could be defined and only a kinetic problem was identified. This could be tackled by transferring the process toward continuous operation, avoiding the slow growth of syntrophic acetate oxidisers once steady state is achieved. The process was modelled using two different methods, where issues surrounding the pH description arise. For the models it is essential to extend the simple buffer capacity description in order to reliably simulate the pH dynamics of the system. As of now information around the microbial community is scarce making the modelling of the alkaline ADM1 a difficult task. To improve the alkaline ADM1 work should be done to determine kinetics of the microbial community and a better description of the substrate with
inoculum. The Dry matter (DM) of the process was 0.88%, which is quite low as in full-scale system usually at least 5% DM is used. The low DM used will lower the chance of bottle-necks in the process, thus with an increase of almost 6 times in DM the inhibition threshold of 420 mgNH4−N will be exceeded. Other implication that need to be solved is the highly saline and high pH waste stream after solids removal. Either this should be recycled back for subsequent digestions or added to the influent of an WWTP assuming it will be diluted enough to not have a major impact anymore. In the future continuous operation should be evaluated as an alternative strategy to prevent the syntrophs from delaying the process and operational parameters like the hydraulic retention time (HRT) and solid retention time (SRT) need to be studied for optimal digestion in such
a system.