MABRs Are Neat, But How Do I Design Them? A Practical Design Methodology for Hybrid MABR/AS

Membrane Aerated Biofilm Reactors (MABR) have been applied at full-scale wastewater treatment plants since 2017 (Underwood et al., 2018) to intensify conventional activated sludge (CAS) to increase flow and/or meet stricter effluent limits. MABRs are typically installed in a pre-anoxic zone and add nitrifying biomass to increase the nitrification capacity (Houweling et al., 2017). The design of MABRs is often done using a combination of process modelling, spreadsheet mass balances, and empirical methods. The objective of this paper is to demonstrate a straightforward design methodology for MABRs that references common design practice for CAS systems.

To achieve reliable nitrification in CAS, process designers typically define a design aerobic solids retention time (aSRTdesign) by applying a safety factor (SF) to a minimum aSRT (aSRTmin) as follows.

aSRTdesign=SF x aSRTmin

Where aSRTmin is commonly defined by kinetic equations presented in Metcalf & Eddy (2014) and can be adjusted to target specific effluent ammonia concentrations across a range of design temperatures, as discussed in Chapter 1 of Houweling and Daigger (2020). Figure 1 shows representative curves for aSRTmin and aSRTdesign with a safety factor of 2 as a function of temperature for an effluent ammonia target of 1 mg/L.

In order to intensify the treatment capacity of nitrifying CAS plants, a media-supported biofilm can be added into the system to enable reliable nitrification at shorter aSRTs than would be typically used. Houweling and Daigger (2020) published a set of design curves relating the fraction of inlet ammonia treated by the biofilm, FNit,B, to aSRT and nitrification safety factor. Figure 2 shows an example of one of these curves at 15°C.

A design methodology is proposed using a combination of the theory used in Figures 1 and 2, where the designer
a) selects a safety factor and design aSRTdesign for a CAS plant to achieve their nitrification goal and,
b) correlates the aSRTdesign and safety factor to a lower aSRT (typically the aSRT available at an overloaded CAS plant, aSRTavailable) plus a fraction of influent ammonia removal by the biofilm.

Application of the design methodology is demonstrated using examples from two real-world MABR designs, with one operating MABR equipped plant and one plant that requires an upgrade using MABR.

Case Study 1
The Hespeler WWTP located in Cambridge, Ontario, Canada is a 9.32 MLD facility that was upgraded with MABR in 2022. The MABR upgrade has achieved year-round nitrification at this facility, which was previously only able to nitrify seasonally (Lakshminarasimman et al., 2023). The design criteria for the Hespeler WWTP was to achieve an effluent of 5 mg/L as ammonia nitrogen at a temperature of 10°C with as low as 4 to 5 day aSRT in the aeration tanks (Natvik et al., 2020).

Figure 3 shows the conventional nitrification curves for the Hespeler case, targeting an effluent ammonia concentration of 5 mgN/L. At 10°C the aSRTmin to achieve 5 mgN/L is 5 days, whereas the aSRTdesign with 1.5 times safety factor is 7.5 days.

Applying the design points to the biofilm safety factor curves by Houweling and Daigger as shown in Figure 4, a 7.5 day aSRT corresponds approximately to a SF of 1.5 when FNit,B = 0. To maintain the nitrification safety factor at 1.5, some removal on the biofilm is required. In this case a 5 day aSRTavailable at SF of 1.5 corresponds to an FNit,B of 0.42, meaning that 42% of the influent ammonia must be removed in the biofilm.

The Hespeler WWTP has 36 ZeeLung™ MABR cassettes by Veolia with 1,920 m2 of surface area each for a total of 69,120 m2. At an average nitrification rate of 1.7 gN/m2/d the MABR will remove 117.5 kgN/d. The influent design ammonia load is 260 kgN/d. Therefore, the MABR will remove 45% of the influent ammonia load, which approximately corresponds to the design points illustrated in Figures 3 and 4.

Case Study 2
A wastewater plant in Texas, USA is approaching design capacity and requires a biological upgrade within a short time frame. The plant has a design temperature of 20°C and is required to meet a monthly effluent ammonia limit of 1.4 mgN/L and can operate at a bulk aSRTavailable of 4.3 days. Figure 5 shows the conventional nitrification curves for the Texas case for a target effluent ammonia of 1.4 mgN/L at 20°C where aSRTmin is 2.8 days and aSRTdesign is 5.5 days with a SF of 2.

Applying a 5.5 day aSRTdesign to the biofilm safety factor curves at 20&Deg;C corresponds to a SF of ~2.25, as shown in Figure 6. In order to maintain a SF of 2.25 at an aSRTavailable of 4.3 days an FNit,B of 0.29 is required.

The influent ammonia load for the Texas WWTP is 1340 kgN/d. To achieve a 29% ammonia removal, 389 kgN/d are required to be removed by the MABR biofilm. Assuming a nitrification rate of 2.6 gN/m2/d, a surface area of ~150,000 m2 is required.

The designs in case studies 1 and 2 use different safety factors, where a less stringent ammonia limit requires a lower safety factor and vice versa. Recommendations for safety factor for ranges of effluent ammonia limits will be presented along with one or more additional case studies.