Treatment of palm oil mill effluent using modified rotating biological contactor with organic loading rate variations

Indonesia is the world’s largest producer of palm oil or Crude Palm Oil (CPO). Palm oil production generates Palm Oil Mill Effluent (POME), which can harm the environment, if not properly treated. Untreated POME, with high Chemical Oxygen Demand (COD) levels, can lead to environmental degradations. Its high organic content and low acidity levels can also cause water pollution. This study modified the Rotating Biological Contactor (RBC) unit to treat POME, increasing the contact surface area between POME and microorganisms. It aimed to analyse the performance of the modified RBC unit, based on variations in organic loading rate (OLR). The study was conducted at a laboratory scale, with 2, 5, and 7 kgCOD/L/day organic loads. The results of this study revealed the highest percentage of removal obtained at an OLR variation of 7 kgCOD/L/day, with a removal rate of 70% for chemical oxygen demand (COD) concentration and 57% for ammonia concentration. Based on the research, the COD removal for organic loads of 7 kgCOD/L/day and 5 kgCOD/L/day decreased to 1900 mg/L and 2600 mg/L, respectively. Thus, the modified RBC unit would be more efficient with an organic load of 7 kgCOD/L/day. However, the COD concentration still did not comply with the water quality standards. Therefore, integrating the modified RBC unit with pre-treatment units, such as coagulation-flocculation, is necessary to achieve optimal effluent concentrations.


Introduction
Indonesia is the largest palm oil producer in the world [1].Based on data from the Indonesian Central Statistics Agency (BPS), the production volume has increased from 2016 to 2021.The production volume of Crude Palm Oil (CPO) reached 31.49 million tonnes in 2016 and increased to 45.1 million tonnes in 2021.The production of CPO has caused environmental issues, such as Palm Oil Mill Effluent (POME).The process of extracting CPO generates a large amount of POME, with a volume ranging from 0.5 to 0.75 m 3 tonnes of processed fresh palm fruit bunches [2].POME has Chemical Oxygen Demand (COD), Total Suspended Solids (TSS), and pH contents of 57,000 -60,400 mg/L, 230 -5,440 mg/L, and 4.65 -4.98, respectively [3].The environmental quality standard, based on the information by the Ministry of Environment and Forestry Regulation No. 5 of 2014, regarding High-Concentration Wastewater Quality Standard, sets the limits for COD, TSS, and oil and grease parameters at 350 mg/L, 250 mg/L and 25 mg/L, respectively.Therefore, POME requires treatment processes to reduce hazardous contaminants, before being discharged into the environment or water bodies, ensuring compliance with the standard quality requirements.
Biological wastewater treatments use microorganisms to degrade organic compounds, aerobically and anaerobically.The commonly used method for treating POME is the aerated lagoon.However, this method requires a large land area and consumes high energy.Therefore, modifications to wastewater treatment technologies are continuously being developed to enhance removal efficiency, with a smaller land footprint and relatively shorter processing time for the POME treatment.
One of the biological treatments implemented using attached growth is the Rotating Biological Contactor (RBC) unit [4].The RBC is designed with a series of disks mounted on a horizontal shaft.Microorganisms grow on the surface of the rotating disks, forming a biofilm layer.The advantages of this unit include a high concentration of active sludge, more optimal sludge settling, low power requirements, and cost-effective operation and maintenance [5].However, the attachment media for microorganisms in conventional RBCs is only located on the outermost surface of the disks, limiting the contact time between wastewater and microorganisms.This results in limited dissolved oxygen (DO) levels and a lower treatment efficiency [6].Efforts to address this issue can be made by modifying the disc plates of the RBC with bioballs, as the filling media.This modification will increase the surface area available for the microorganism growth, thereby supporting improved efficiency in removing Biochemical Oxygen Demand (BOD), COD, and Ammonia.
The performance of the RBC is influenced by the rotational speed, hydraulic retention time (HRT), organic loading rate (OLR), rotational ratio, media submergence depth, and dissolved oxygen (DO) levels [7].This study's OLR is the main focus as the research limitation.OLR indicates the amount of organic compound in wastewater that microorganisms can biologically degrade a reactor per unit volume per day [8].The organic material being degraded is represented by the COD parameter.The typical range of OLR values in conventional RBC units for treating organic materials with a COD parameter is between 600 and 2,000 mg/L [9].Increasing OLR can lead to an excessive biofilm production and decreased dissolved oxygen (DO) levels [10].At a high OLR (Organic Loading Rate), there is typically an increase in the supply of nutrients and oxygen to the RBC unit to meet the higher demands of microorganisms.With sufficient availability of nutrients and oxygen, microorganisms can function optimally to degrade organic materials, ultimately enhancing COD removal efficiency [11].The modification of the RBC unit aims to increase the OLR by enhancing the contact area between wastewater and microorganisms.Changing the attached media using bioballs also allows more biochemical reactions to occur, thereby increasing the efficiency of COD removal [12].Based on previous research, the RBC reactor treated tofu liquid waste with an organic load of 6,420 mg/L, reaching 88% with a detention time of 24 hours [13].However, because the characteristics of POME are higher of organic compound, a combination of attached growth and suspended growth processes is carried out.This research aims to analyse the treatment performance and biokinetic model, based on variations in the OLR.

Methods
The research samples were obtained from the POME of PT Perkebunan Nusantara (PTPN) VIII, Cigudeg, Bogor Regency, following the sampling standards outlined in SNI 6989.59:2008 on Wastewater Sampling Methods.The study was conducted at a laboratory scale in the Water Quality Laboratory, Department of Civil and Environmental Engineering, IPB University.

Tools and materials
The RBC for aerobic treatment is made of acrylic with a capacity of 14.5 L and equipped with a 15 cm diameter frame.Wire, frames, and acrylic cylinders are used to assemble the stages and trains of the RBC.Bioball media fills the RBC disks, a total of 150 pieces per unit.Additionally, a motor gearbox and time delayer are used to regulate the rotational speed of the RBC.Adjacent to the motor gearbox, a small fan is installed to reduce the temperature around the motor gearbox.
The use of equipment and chemical substances for water quality testing to assess reactor performance is based on the parameters to be analysed, which include the COD analysis, using the method specified in SNI 6989:02:2019, which involves closed reflux spectrophotometry, Ammonia analysis using the method described in SNI 6989:30:2005, which employs spectrophotometry, based on the Nessler method, pH measurement using the method outlined in SNI 6989:11:2019, which provides guidelines for acidity testing, Mixed Liquor Suspended Solids (MLSS) analysis using the method standard in SNI 6989:03:2019, which includes gravimetric determination of TSS and Total Dissolved Solids (TDS) analysis.These methods utilise specific equipment and chemical reagents, as required by the respective standards for accurate analyses.

Research procedure
The research activities are divided into 5 stages; reactor preparation, reactor configuration, seeding and acclimatisation, main research sampling, and wastewater quality analysis.The reactor operation is conducted based on the influence of OLR variations.The study focuses on three OLR scenarios: 2,000, 5,000, and 7,000 mgCOD/L/day.A conventional RBC can treat organic waste, with COD parameters ranging from 600 to 2,000 mg/L [9].For each organic load variation, samples are taken daily for 1 week.Replacing discs in the conventional RBC with bioballs in the modified RBC is expected to treat wastewater with higher organic loads.The hydraulic retention time is kept constant for one day, resulting in a pump flow rate of 0.6 L/hour for a unit volume of 14.5 L. The relatively long hydraulic retention time is necessary for the RBC unit, due to the high organic load carried by the POME.Therefore, the reactor operation combines attached growth and suspended growth types with a 40% submergence of the suspended media.Sampling is conducted at 2 points in the influent and effluent of the modified RBC unit for wastewater quality testing, based on COD, TSS, and pH parameters.
The seeding process or microbial propagation is carried out to increase the microbial population.The seeding process involves the continuous addition of glucose into the RBC reactor, conducted for 20 days, during which, a biofilm layer will form and coat the bioballs.This process is followed by acclimatisation, which takes place over 64 days.The transition from glucose to POME is divided into 4 stages.Acclimatisation stage 1 has a ratio of 25:75 and is intended to achieve a target COD of 2,000 mg/L.Stage 2, with a ratio of 50:50, is planned to meet a target COD of 3,000 mg/L.Acclimatisation stage 3 with a ratio of 75:25 is targeted to achieve a COD of 4,500 mg/L.The acclimatisation process ends when 100% of the glucose is replaced with POME waste at a ratio of 100:0, targeting a COD of 5,000 mg/L.The biofilm layer on the bioballs will thicken as the microorganisms attached to the bioballs become ready for use.Based on the microorganism growth medium, biological treatment processes are classified into attached growth and suspended growth.In an attached growth system, microorganisms have more opportunities to interact with organic matter in wastewater, accelerating the degradation and removal of pollutants [14].In this biofilm layer, microorganisms form tightly bound colonies between themselves and the media surface, creating a strongly adhered matrix.This biofilm forms a complex microbial environment and provides an ideal habitat for microorganisms to interact, exchange nutrients, and perform biochemical reactions that support wastewater treatment processes [15].As for the suspended media, the biomass is spread homogeneously in the supernatant, so the concentration of MLSS in the suspended media tends to be higher [16].

Seeding and acclimatisation
Seeding is carried out to cultivate microorganisms, by obtaining microorganisms from a source with good degradation capabilities.The seeding process takes 20 days, until the Mixed Liquor Suspended Solids (MLSS) concentration reaches 10,000 mg/L and the removal of high COD concentration becomes stable, after which, acclimatisation will take place.The highest decrease in COD concentration occurs on day 11, with a reduction of 90% within one day (Figure 2).The highest inlet COD concentration is recorded on day 11 at 4,971 mg/L, decreasing by 47% on day 12 to reach 2,644 mg/L (Figure 2).The total COD concentration reduction during the seeding process is the highest, reaching 95%, from an initial COD concentration of 4,971 mg/L to 257 mg/L.The first acclimatisation stage lasted for 16 days, and the target COD concentration of 2,000 mg/L was achieved since day one.The second acclimatisation stage took 16 days, and the target COD concentration of 3,000 mg/L was met on day 26.The third acclimatisation stage lasted for 20 days, and the target COD concentration of 4,500 mg/L was achieved on day 50.The final acclimatisation stage lasted 12 days, and the target COD concentration of 5,000 mg/L was fulfilled on day 61.During the acclimatisation process, the addition of substrate caused the microorganisms to readjust, increasing the COD concentration.Besides the substrate, fluctuations in COD, as shown in Figure 2, act as a carbon source for microorganisms due to several other factors, such as temperature, pH, and nutrient sources [17].

Analysis of palm oil mill effluent quality
The characteristics of the POME at PTPN VIII Cigudeg, Bogor Regency, based on the concentration of COD, ammonia, TDS, TSS, and pH, respectively, are 20,804.92mg/L, 3.92 mg/L, 1,984 mg/L, 70,945 mg/L, and 3.8, respectively (Table 1).According to the Regulation of the Ministry of Environment of the Republic of Indonesia, Number 5, Year 2014 Attachment III, on the Standard for Wastewater Quality for Palm Oil Industry, all measured parameters have values that exceed the standard limits.RBC can treat COD and TSS concentrations.However, there are limitations to high concentrations.
Increasing the organic load in wastewater can lead to a decrease in the degradation capacity of the wastewater treatment reactor [20].However, the efficiency in reducing the pollution level in wastewater will increase with the increase in OLR, until a certain point is reached [21].

Effect of OLR on RBC
The COD removal efficiency in the modified RBC reactor operated at a laboratory scale based on the OLR variations of 2000, 5000, and 7000 mg COD/L/day ranged from 34% to 62%, 51% to 65%, and 65% to 70%, respectively.Based on Table 2, the influent COD concentrations were in the range of 4,970 mg/L to 5,428 mg/L, 7,128 mg/L to 7,701 mg/L and 2,461 mg/L to 2,837 mg/L, for the first variation, the second variation, and the third variation, respectively.The influent COD concentrations tend to fluctuate, due to the changing characteristics of POME caused by biodegradation processes, during sampling.The COD, TSS, and pH parameters were analysed for 21 days to determine the effect of the modified RBC unit on the contaminant removal efficiency.Based on the performance of the modified RBC, the effluent COD concentration for an OLR of 5,000 mg/COD/L/day had an average of 2,180 mg/L, while for an OLR of 7,000 mg COD/L/day, it had an average of 2,399 mg/L.Therefore, an OLR of 7,000 mg COD/L/day was better than an OLR of 5,000 mg COD/L/day, because it resulted in similar effluent concentrations, despite different organic loads.The highest percentage of COD removal, at 70%, was achieved at an OLR of 7,000 mg COD/L/day.This proves that, processing will be more optimal at a larger organic load.This study also produced a percentage of COD parameter removal similar to that found in a previous study by Angga, 2014 [13].The bioballs in the modified RBC unit provide a significant surface area expansion.This larger surface area provides more space for microorganisms involved in decomposing organic materials from POME.With a high Organic Loading Rate (OLR) of 7,000 mg COD/L/day, the amount of organic matter entering the RBC unit increases, and the presence of bioballs with a large surface area helps enhance the contact between organic materials and the existing microorganisms.For an OLR of 2,000 mg COD/L/day, the effluent COD concentration averaged 1,265 mg/L.This result indicates that, the treatment is more optimal at higher organic loads.However, the effluent concentrations in each variation did not meet the wastewater quality standard stated in the Regulation of the Ministry of Environment and Forestry of Indonesia No. 5/2014, regarding Wastewater Quality Standards.Thus, further treatments are required to achieve effluent concentrations below the required standards.
The microorganisms within the modified RBC unit experience both growth and mortality.Based on the information in Figure 3, the concentration of MLSS attached to the bioballs ranged from 1,307 mg to 1,825 mg, with an average of 1,614 mg.According to Table 2, the TSS concentration in the modified RBC ranged from 10,000 mg/L to 16,800 mg/L, with an average of 13,190 mg/L.The TSS concentration can decrease, if the death rate of microorganisms is higher than their growth rate.The removal of TSS in the modified RBC unit ranged from 93% to 98% (Table 2).The influent and effluent TSS concentrations in the modified RBC were in the range of 3,600 mg/L to 5,400 mg/L, 4,400 mg/L to 8,800 mg/L and 2,800 mg/L to 4,200 mg/L for the first variation, the second variation and the third variation, respectively.The effluent concentrations for the first and second variations ranged from 200 mg/L to 400 mg/L, while the third variation had an average of 200 mg/L.
The POME treatment process with modified RBC is carried out at the room temperature, and as such, there is no specific treatment to stabilise the pH and temperature.The pH values in the influent were in the range of 3.47 -4.53, 3.13 -3.64 and 4.26 -4.53, for the first, the second, and the third variations, respectively.Meanwhile, the pH values in the effluent were in the range of 5.08 -6.17, 4.68 -5.36 and 5.07 -6.21, for the first, the second, and the third variations, respectively (Figure 4).The optimum pH for microorganisms to function in aerobic conditions is 6 -8 [22].The pH conditions in the effluent for the first variation only met the quality standard on day 5.For the second variation, none of the pH values in the effluent met the quality standard.The third variation, from day 4 to day 7, met the quality standard, although only marginally.It also demonstrated that, microorganisms can be more efficient in treating wastewater with an optimal pH.The optimum pH value for aerobic biological treatment microorganisms is 6 -8 [22].Ca(OH)2 + water can be added to neutralise the pH for pH concentrations, that do not meet the quality standard.These results demonstrate the influence of the organic load on the modified RBC unit on the COD and TSS removal efficiency.

Conclusion
The use of bioballs as the attached media in the modified RBC unit has a positive impact, because it can effectively treat wastewater with a high organic load.The surface area of bioballs provides a larger growth area for microorganisms, compared to conventional RBC systems.Based on the performance results of the modified RBC unit, the optimum COD concentration at an OLR of 7,000 mg COD/L/day is in the range of 2,223-2,591 mg/L.However, these results still did not comply with the effluent quality standards set by the Ministry of Environment and Forestry of the Republic of Indonesia Regulation No. 5/2014, for Palm Oil Industry Wastewater, which sets the limit at 350 mg/L.Therefore, the modified RBC unit must be integrated with pre-treatment units, such as coagulation-flocculation, to achieve the desired effluent quality.

Figure 1 .
Figure 1.RBC modified reactor: (a) physical condition of the unit during operation, (b) formed biofilm layers.

Figure 2 .
Figure 2. Performance of the RBC modification unit based on OLR variations on the parameters COD.

Figure 3 .
Figure 3.The performance of the RBC modification unit based on OLR variations on the parameters.TSS.

Figure 4 .
Figure 4.The performance of the RBC modification unit based on OLR variations on the parameters pH.

Table 1 .
Comparison of POME characteristics with quality standards.

Table 2 .
COD concentration on variations in OLR in RBC.