Enhanced Microbial Treatment of Garlic Wastewater Using MBBR + A/O Process
Overview
Garlic wastewater primarily originates from slicing and rinsing processes during garlic processing. It is characterized by high concentrations of organic matter, significant levels of nitrogen and phosphorus, and contains substantial amounts of allicin. Allicin (diallyl thiosulfinate) is a volatile liquid responsible for the pungent odor of garlic and is chemically unstable and highly reactive. Allicin can inhibit the growth of various microorganisms. Discharging high-concentration garlic wastewater without treatment causes serious environmental impacts. Some researchers have employed techniques such as membrane filtration, Fenton oxidation, and micro-electrolysis, but these methods have not been effective for treating garlic wastewater, and the use of large doses of chemicals increases subsequent treatment costs. Many scholars have proposed biological treatment methods using anaerobic-aerobic combined processes. However, due to the antibacterial properties of allicin, microorganisms are difficult to cultivate, and the treatment efficiency is not ideal. Therefore, the focus of biological treatment is to cultivate and acclimatize microbial strains capable of adapting to garlic wastewater and to enhance their biodegradation.
This study involved cultivating and screening bacterial strains effective in degrading garlic wastewater, which were then introduced into a Moving Bed Biofilm Reactor (MBBR). Using inoculated sludge and a flow-rate increasing biofilm formation method, biofilms were established to enhance the removal of nitrogen and phosphorus from the wastewater. This was followed by further A/O (Anoxic/Oxic) biochemical treatment. According to the GB18918-2002 standard, the effluent COD and Ammonia Nitrogen (NH₃-N) levels can meet the secondary standard (COD: 100 mg/L, NH₃-N: 25-30 mg/L). This process effectively reduces the organic content in the effluent, lowering the difficulty of subsequent treatment stages.
1. Experimental Section
1.1 Process Flow Design
The overall process flow for garlic wastewater treatment is shown in Figure 1, with the core component being the biodegradation in the MBBR + A/O system. Three screened and isolated strains effective in degrading garlic wastewater – Alcaligenes sp., Acinetobacter sp., and Achromobacter sp. – were mixed with activated sludge and introduced into the MBBR unit to facilitate its rapid start-up.
1.2 MBBR + A/O Treatment Process
After passing through coarse and fine screens to remove suspended solids, the garlic wastewater is pumped directly into the MBBR. The influent quality is shown in Table 1. The effluent from the MBBR flows directly into the A/O system. Due to the low organic content of the MBBR effluent, raw garlic wastewater is appropriately added to the Oxic (O) tank to supplement the carbon source for the A/O process. To test the system's impact resistance, the organic loading rate of the MBBR was gradually increased during continuous operation, and the effluent quality was monitored.
1.3 Process Parameters
1.3.1 Dissolved Oxygen (DO)
Excessively high DO within the biofilm can prevent denitrification, causing the MBBR to lose its simultaneous nitrification and denitrification capability. Excessively low DO can lead to the proliferation of filamentous bacteria, affecting effluent quality and inhibiting the nitrification process.
1.3.2 Hydraulic Retention Time (HRT)
An excessively short HRT causes intense reaction conditions, where wastewater containing most organic matter is discharged before being fully absorbed. Continuous inflow keeps microorganisms in a constant state of biodegradation, reducing efficiency and increasing energy consumption. An excessively long HRT leads to depletion of nutrients; without nutrients, microorganisms reduce their activity and metabolic demands to merely sustain survival.
1.3.3 Carbon-to-Nitrogen Ratio (C/N)
A low C/N ratio can lead to the catalysis of ammonia conversion into other substances, affecting ammonia nitrogen removal. It also easily causes filamentous bulking, continuous growth affecting flocculation, leading to sludge bulking and floating sludge. A high C/N ratio is unfavorable for microbial biodegradation and growth, increasing the organic load on the microorganisms.
1.4 MBBR Biofilm Start-up
Biofilm Start-up: The inoculated sludge + flow-rate increasing method was used. MBR-enriched activated sludge was inoculated into the reactor, with an initial Mixed Liquor Suspended Solids (MLSS) concentration of approximately 5.82 g/L. Aeration was started, and polyethylene carriers were added to the reactor with a filling ratio of about 60%. The DO in the reactor was controlled above 4.0 mg/L. The influent flow rate was increased stepwise in 20 L/h increments: 20, 40, 60, 80, 100, 120, 140 L/h, with each flow rate maintained for 1 day. No sludge was wasted during this phase. A light yellow biofilm formed on the surface of the carriers where microorganisms attached and grew. After successful biofilm start-up, stable operation continued, maintaining a Sludge Retention Time (SRT) of 30 days. During stable operation, the organic loading rate of the MBBR was adjusted to observe its impact on COD, nitrogen, and phosphorus removal.
2. Results and Discussion
2.1 Analysis of MBBR Effluent Quality During Biofilm Start-up
The aeration intensity in the MBBR was adjusted to control the DO concentration. When DO was below 4.0 mg/L, the aeration intensity was insufficient to support uniform, high-flow turbulent motion of the carriers, preventing adequate mixing and making it difficult to form a biofilm on the carrier surfaces. When DO was between 4.0–6.0 mg/L, the carriers mixed thoroughly with the activated sludge and wastewater. A color change from white to yellowish-brown on the carriers was observed, indicating successful microbial attachment and growth under this aeration intensity, as shown in Figure 2.
The variation curve of influent and effluent COD during the start-up phase is shown in Figure 3(a). The initial decrease in treatment efficiency was due to the very low quantity of attached microorganisms on the carriers; the degradation by microorganisms in the activated sludge alone was insufficient to remove the large amount of organics. As the start-up progressed, the amount of attached microorganisms on the carriers increased, gradually forming a biofilm. The effluent COD concentration gradually stabilized, and COD removal efficiency stabilized above 90%.
The variation curve of MBBR influent and effluent NH₃-N is shown in Figure 3(b). The nitrification by aerobic bacteria in the activated sludge effectively removed ammonia nitrogen. Starting from day 7, the influent NH₃-N concentration gradually increased. By day 23, although the influent NH₃-N was still increasing, the removal rate also increased. This was because nitrifying bacteria grow slowly initially; with time, their population increased, the biofilm matured, and the NH₃-N removal rate gradually increased and stabilized.
The variation curve of MBBR influent and effluent TN is shown in Figure 3(c). Unlike ammonia nitrogen removal, the TN removal efficiency decreased initially. This was because the reactor environment had ample oxygen and carbon source, limiting the growth of denitrifying bacteria. However, as the biofilm formed, TN removal efficiency began to improve. By day 20, although the influent TN concentration increased, the effluent TN and removal rate stabilized, ranging between 50%–60%.
The variation curve of MBBR influent and effluent TP is shown in Figure 3(d). From start-up to stable operation, the TP removal rate remained stable. Although the influent TP concentration was high initially and decreased later, the removal efficiency showed no significant change, indicating the system's capability for phosphorus removal. The TP removal rate in the system was maintained between 80%–90%.
In summary, maintaining the MBBR system DO between 4–6 mg/L, a mature biofilm developed after 20 days of continuous feeding. Compared to traditional activated sludge processes, the MBBR system offers strong impact resistance and high treatment efficiency, effectively reducing the difficulty of subsequent treatment stages for garlic processing wastewater.
2.2 Effluent Quality Analysis During Stable Operation
After the biofilm start-up phase, the biofilm matured. To test the impact resistance of the MBBR system, the organic loading rate was continuously increased during stable operation.
The variation curve of MBBR influent and effluent COD during stable operation is shown in Figure 4(a). From days 1–5, with constant inflow, COD removal efficiency remained above 95%, and effluent COD concentration reached around 100 mg/L. From days 5–20, the inflow rate was increased, gradually raising the organic loading from 20 kgCOD/m³·d to 30 kgCOD/m³·d. No significant change in removal efficiency was observed, and effluent COD remained between 80–100 mg/L, demonstrating strong impact resistance. After day 20, the inflow rate was further increased, continuously raising the organic loading in the reactor from 30 kgCOD/m³·d to 37 kgCOD/m³·d, maintained for 5 days. The MBBR's COD removal capacity remained above 95%.
Figures 4(b) and (c) show the variation curves for NH₃-N and TN, respectively, during stable operation. From days 1–5, with constant inflow, the MBBR biofilm exhibited simultaneous nitrification and denitrification. Aerobic nitrifying bacteria attached to the outer layer of the biofilm, fully mixed with the wastewater under aeration, consumed significant nitrogen sources through nitrification. Denitrifying bacteria in the inner anoxic layer effectively removed nitrate nitrogen through denitrification. From days 5–20, as the inflow rate increased, the removal efficiency of NH₃-N and TN initially decreased significantly. After about 7 days of continuous operation, the system gradually adapted. Although the removal efficiency for NH₃-N and TN then increased, it remained lower than during the low-flow period. Under constant inflow, NH₃-N removal reached over 90%, with effluent NH₃-N between 10–15 mg/L, and TN removal was basically maintained above 80%, with effluent TN around 30 mg/L. After increasing the inflow and the system reached a new balance under continuous impact, NH₃-N removal stabilized around 80%, with effluent NH₃-N between 50–70 mg/L, and TN removal around 60%, with effluent TN below 50 mg/L.
The variation curve for TP during stable operation is shown in Figure 4(d). The effluent TP concentration was basically maintained around 10 mg/L. Initially, with constant low flow and low influent TP concentration, the treatment effect was limited. As the inflow rate and influent TP concentration increased, high treatment efficiency was achieved throughout the impact phase and the subsequent high-load operation, with the TP removal rate fluctuating around 90%.
In summary, under high organic loading shock, the system's COD removal efficiency remained largely unchanged, but the removal of NH₃-N and TN decreased more significantly. When the organic loading reached its maximum of 37 kgCOD/m³·d, the system's removal efficiency for NH₃-N and TN decreased noticeably.
2.3 Effluent Quality Analysis of MBBR + A/O System
After the biofilm start-up phase and one month of stable operation, an A/O process was added downstream for advanced treatment of the MBBR effluent. Gradient increases in inflow rate were applied to increase the overall organic loading, aiming to determine the optimal inflow rate, corresponding to the optimal HRT.
The COD variation curve is shown in Figure 5(a). The inflow rate increased sequentially: 100, 120, 130, 150, 170 L/h. From the start to the maximum flow rate, the organic loading on the MBBR system increased from 20 kgCOD/m³·d to 37 kgCOD/m³·d. The final effluent from the combined system remained stable, with COD concentration below 100 mg/L. Under sustained high organic loading shock, the MBBR system performed well, although its effluent COD showed a slight increase when the flow rate reached 150 L/h. After maintaining the 170 L/h flow rate for several days, a noticeable upward trend in the MBBR effluent COD was observed. However, with the subsequent A/O process, the final combined system effluent was still maintained below 100 mg/L. This indicates that even under the high organic loading shock of 37 kgCOD/m³·d, the combined process still has a strong removal effect on garlic processing wastewater.
The variation curves for NH₃-N and TN are shown in Figures 5(b) and (c), respectively. Garlic processing wastewater has high concentrations of ammonia nitrogen and total nitrogen, which can further increase over time due to oxidation. Typically, ammonia nitrogen concentration ranges from 300–500 mg/L, and total nitrogen from 450–600 mg/L. Under the simultaneous nitrification and denitrification in the MBBR, ammonia nitrogen removal was more effective, likely because nitrifying bacteria utilize the wastewater more efficiently under aeration. Denitrifying bacteria require anoxic conditions and often depend on consumed organic carbon for denitrification. When increasing the inflow rate, the removal efficiency of NH₃-N and TN was the primary consideration. From days 1–4, due to low flow rate and moderate NH₃-N, the NH₃-N removal rate remained above 90%, and TN removal efficiency gradually increased. Subsequently, the inflow rate was significantly increased. It was clearly observed that as the inflow rate increased, the effluent concentrations of NH₃-N and TN at different stages sequentially rose, with higher inflow rates leading to higher effluent concentrations. As the flow rate increased, the biomass on the biofilm carriers increased, enhancing nitrification, where ammonia nitrogen is oxidized by nitrifying bacteria to nitrate and nitrite under oxygen.
The TP concentration variation curve is shown in Figure 5(d). Given the high influent COD and TN concentrations, the theoretical optimal TP concentration for microbial growth is above 100 mg/L. However, the influent TP concentration was far below this theoretical requirement. Therefore, the MBBR effluent TP concentration remained around 10 mg/L, and the final combined system effluent TP concentration was maintained between 2–3 mg/L.
The sludge characteristics of the MBBR system and the subsequent A/O system before and after operation were measured, as shown in Table 2.
In summary, when the flow rate was increased to 150 L/h, the removal rates for COD, NH₃-N, TN, and TP were superior to those at other flow rates. The HRT at this flow rate was 27 hours. Furthermore, the sludge concentration in both the MBBR and A/O systems increased substantially after operation.
3. Conclusion
After biofilm formation in the MBBR, the removal efficiencies for COD, NH₃-N, TN, and TP were stable. During one month of continuous operation under stable conditions, COD removal reached over 95%, NH₃-N and TN removal stabilized around 80%, and TP removal stabilized around 90%.
The MBBR effluent was further treated in the A/O system. The combined process could withstand an organic loading up to 37 kgCOD/m³·d. The optimal operation for the overall process was under an HRT of 27 hours. The final effluent COD stabilized below 100 mg/L, NH₃-N between 10–20 mg/L, TN below 30 mg/L, and TP below 10 mg/L. The sludge concentration in the MBBR system after operation was 8.5 g/L, and in the A/O system was 4.1 g/L, both significantly higher than before operation, indicating a substantial increase in microbial biomass. The COD and ammonia nitrogen levels after biological treatment met the secondary discharge standard of GB18918-2002. For further treatment, Fenton's advanced oxidation technology could be employed for deep treatment of the biologically treated effluent to achieve the first-level discharge standard.