Effect of the co-digestion of agricultural lignocellulosic residues with manure from South American camelids

This study aims to evaluate the effects of the co-digestion of agricultural residues with manure from camelids from the Andean zone. Different combinations of llama manure (LM) and vicunas (VM) were made with amaranth (AS), quinoa (QS), and wheat (WS) residues. They were fermented using sewage sludge as inoculum. The co-digestion was evaluated under mesophilic conditions for 40 days. The ratios of volatile substances of substrate / co-substrate evaluated were 0:100; 25:75; 50:50, 75:25, and 100:0. Two substrate / inoculum ratios (SIR 1:1 and SIR1:2) were also evaluated. The results indicate that the maximum methane accumulation rate is obtained in SIR 1:1 for a VM-AS ratio (25:75) with 540 mL/g volatile solid (VS). In general, the results did not increase with the increase in inoculum; rather, the tendency to improve methane yield is associated with an increase in the amount of agricultural residues, mainly AS. Regarding the kinetic modeling, the transfer model is the one that best adjusted the predicted values to those observed with an r(2) between 0.991 and 0.999, and an RMSE value between 2.06 and 13.62 mL/g (volatile solid) VS. Finally, all the trials presented synergistic effects in their co-digestion except the digesters formed by LM-AS, LM-QS and LM-WS of SIR 1:2. These presented antagonistic effects in which the addition of the co-substrate generated competition with the substrates, reducing methane production. (c) 2021 Society of Chemical Industry and John Wiley & Sons, Ltd

Automated summary

The study evaluated the co-digestion of agricultural residues with camelid manure in the Andean zone, finding that the highest methane accumulation rate was obtained with a VM-AS ratio (25:75) under a SIR 1:1 condition. Overall, increasing the inoculum did not increase methane yield, but rather the amount of agricultural residues was associated with improved methane production. The kinetic modeling showed that the transfer model best adjusted predicted values to observed values.