Assessment on AnMBR for Removing Lipid and Carbohydrate from POME

Palm oil is a popular ingredient in domestic products. The palm oil industry has been growing rapidly over the past decades, so that the amount of palm oil mill effluent (POME) generated from the palm oil production has been increasing as well. The anaerobic membrane bioreactor (AnMBR) is a treatment solution that can remove organic pollutants from POME while generating methane as an energy source. In comparison to conventional anaerobic digestors, the AnMBR technology has an additional membrane unit that can produce effluent with higher water quality. More specifically, if ultrafiltration is applied, the AnMBR will be able to effectively remove bacteria from the effluent, making it suitable for direct fertigation (Uman et al., 2021; Bray et al., 2021). However, in cases where infectious viruses are also present, further disinfection method might be required. In this experiment, a lab-scale AnMBR system was used for POME treatment. In order to evaluate how well the system can perform in terms of pollutant removal and methane production, under the controlled experimental conditions, several criteria were monitored: (1) chemical oxygen demand (COD) removal, (2) biomass growth, (3) biogas production, (4) digestion efficiency, and (5) volatile fatty acids (VFA) accumulation. A Long chain fatty acids (LCFA) analysis method was developed using the liquid chromatography/mass spectrometry (LC/MS), to elaborate on underlying conversion mechanisms. A COD balance analysis was also conducted. Factors that would potentially contribute to the COD gaps in the COD balance analysis were quantified and discussed in this paper as well to validate the experimental results. The solid retention time (SRT) was controlled at 140 days, and the organic loading rate (OLR) at 3 gCOD·L-1·d-1 during the first phase of the experiment, when synthetic POME and VFAs were added to the bioreactor. During the second phase, the SRT and the OLR of POME remained the same, whereas the VFAs were replaced by starch and the OLR of starch was increased, in order to simulate the real POME composition, because in addition to lipid, carbohydrate and protein are also found in POME. During Phase I, the AnMBR system could remove 98%-99% of the incoming COD, and produce about 5 L of methane each day. During Phase II, the microbes did not have enough time to adapt to the new experimental condition, but the stability of the AnMBR system could be achieved overtime, when the mixing is improved and the buffer solution is adjusted properly according to the pH variation. Although, based on the positive biomass net growth and the increased methane production, it could be predicted that adding carbohydrates to the feed for a more representative POME composition would promote biomass growth and methane production, suggesting that the AnMBR system would have higher potential when the real POME is used for energy recovery.Civil Engineerin