Tertiary treatment of palm oil mill effluent using sequential batch reactor

Abstract

Palm oil mill effluent (POME) is a liquid by-product of palm oil processing. Today, most palm oil mills in Malaysia have adopted ponding treatment system as a primary and secondary treatment of POME. However, this system alone is not sufficient in producing treated effluent that can fulfil the effluent discharge standards. This study was therefore conducted to investigate the feasibility of sequencing batch reactor (SBR) as a tertiary treatment for POME. The study started with examining the effectiveness of ponding treatment system at Pertubuhan Peladang Negeri Johor (PPNJ) Kahang Palm Oil Mill in treating POME and the effect of average rainfall, quantity and quality of the processed fresh fruit bunch (FFB) on raw POME characteristics and ponding treatment system performance. This was followed by the assessment of SBR performance at both laboratory and pilot-scale. The SBR operations were evaluated during its acclimatisation stage as well as the effects of organic loading rate (OLR) and food over microorganism ratio (F/M) on the SBR treatment performance. It was found that PPNJ ponding treatment system, which comprised of five ponds are efficient for POME treatment. The average removal in terms of Total Chemical Oxygen Demand (TCOD), Soluble Chemical Oxygen Demand (SCOD), Total Nitrogen (TN), Total Phosphorus (TP) and Total Suspended Solids (TSS) was 96%, 96%, 86%, 96% and 95%, respectively. However, even though the removal percentage was high, only pH and TN fulfilled the Department of Environment (DOE) effluent discharge standards, whereas the concentration of TSS and TCOD in the treated effluent exceeded the allowable limit. The ponding treatment system was also incapable to decolourise POME and it increased the concentration of Ammonia-Nitrogen (NH3-N) in the treated effluent. Both quantity and quality of processed FFB affected the colour intensity in POME. The average rainfall has minor effect on TP removal, while the quantity of FFB processed influenced SCOD, TCOD and colour removal by the ponding treatment system. Rainfall variation and quantity of processed FFB were found not to significantly affect the removal of TN, TSS and NH3-N by the ponding treatment system. When the partially digested POME was further treated using laboratoryscale SBR, the removal of SCOD ranged from 5 to about 49% with an average of 27%. The best OLR and F/M ratio for the laboratory-scale SBR were 0.58 kg COD/m3 day and 0.26 g COD/g MLVSS day, respectively. For the pilot-scale SBR, the removal of SCOD ranged from about 3% to 65% with an average of 32%. The best OLR and F/M ratio for the pilot-plant SBR were 0.72 kg COD/m3 day and 0.17 g COD/g MLVSS day, respectively. Comparatively, SCOD removal by the pilotscale SBR was slightly better than the laboratory-scale SBR at the best OLR and F/M ratio. In conclusion, ponding systems had demonstrated to be effective for POME treatment except for NH3-N and colour removal. Further treatment of partially treated POME by SBR yields further organic removal.