Anaerobic co-digestion of agricultural waste for biogas production

Abstract

South Africa today has an issue of high demand for energy, water, and waste management. The utilisation of promising alternatives can be achieved through the sustainable management of agricultural wastes (AW) and their valorisation to produce biogas. This management adds to the fight to move away from the environmentally detrimental fossil fuels. The method of waste management being used nationwide is landfilling which has been found to be a significant risk to the environment since it emits greenhouse gases (such as carbon dioxide and methane). In response to this problem, this study aimed to valorise the AW and market wastewater (MWW) obtained from a local fruit and vegetable bulk market to produce biogas using co-digestion. The study utilised the apples, bananas, oranges, spinach, tomatoes, butternuts, potatoes, and carrots as the AW. The activated sludge (AS) was obtained from a local wastewater treatment facility and was used as the inoculum. The first objective was to characterise the AW to determine the properties that allow application in the anaerobic co-digestion (Co-AD) process. The solid AW was subjected to mechanical pre-treatment to increase the surface area and enhance a better co-digestion application. The analysis was conducted using the Fourier-Transform Infrared Spectroscopy (FTIR), Brunauer–Emmett–Teller (BET), Scanning Electron Microscopy (SEM) and oxygen bomb calorimeter. The second objective investigated the feasibility of the process of Co-AD of different AW, such as apples, bananas, carrots, butternuts, and potatoes, combined with market wastewater (MWW). The Co-AD process used activated sludge (AS) as the inoculum. A biochemical methane potential test (BMP) in 1 L capacity digesters assessed the potential for biogas production. The digesters were fed feedstock with an organic loading rate (OLR) of 2.5 kgVS/m3 .day maintained at a temperature of 40℃ for a hydraulic retention time (HRT) of 21 days. The BMP used a mixing ratio of 1:1 (% w/w) between MWW and AWs and a ratio of 1:2 between the co-substrates and inoculum. The third objective was to develop a predictive model using the Box-Behnken design (BBD) model matrix on the response surface methodology (RSM). The input factors were temperature (32 - 42°C), pH (6 - 8), HRT (10 - 21 days) and OLR (1 - 4 kgVS/m3 day). The final objective was to analyse the kinetics of the Co-AD system through the application of Modified Gompertz and first-order kinetic models. Among the AW, spinach waste was found to have the highest BET surface area of 1.70 m2 . The apple and banana waste developed pore volumes of 0.068 nm and 0.69 nm, respectively The common functional groups included the O-H, C-O and carboxylic acids with wavelengths ranging from 720 to 3280 cm-1 for all the AW. The energy content in the agricultural waste ranges between 14 and 15 MJ/kg. These results confirmed the successful mechanical pretreatment of agricultural waste and its viability in being a feedstock for Co-AD. Among the different substrates, the apple and banana substrates showed the highest biogas output, with 595 mL/day with a methane composition of 68% and 585 mL/day with a methane composition of 65%, respectively. The control digester produced 450 mL/day of biogas, with 60% methane composition. The second stage evaluated the effect of substrate-to-substrate (SS) ratios and the different combinations of agricultural waste. The best-performing combination, Mix-1, produced 680 mL/day at an SS ratio of 2, and it also produced the highest biogas production with a methane composition of 75%. At optimum conditions of OLR of 3.98 kgVS/m3 .day, a temperature of 40°C, HRT of 10 days, a pH of 7.2, a biogas production of 716.53 mL/day, a VS reduction of 73.37% and COD removal of 79.24%, with a desirability of 100% was obtained. The correlation between the predicted results and the experimental data was high, with a correlation coefficient (R2 ) value between 0.9 and 1. The findings from the kinetics analysis indicated that the Modified Gompertz model provided a more accurate description of the kinetics and dynamics of the system compared to the first-order kinetic model. This conclusion was supported by a correlation coefficient (R2 ) exceeding 0.99 and an error margin of less than 2%. Finally, when co-digested with market wastewater, agricultural waste exhibited good performance in biogas production, especially apples, bananas, and Mix-1 combination. The study supports finding new ways to mitigate agricultural waste and protect the environment. Further studies may include more types of agricultural waste and evaluate other co-substrates that are easily obtainable. The study also proposes new energy sources that investigated further that will promote environmental sustainability.