SADp, i.e. SAD based on permeate volume, is varies not only by membrane configuration and module design, but also by MBR system design and operating condition. It tends to be lower for hollow fiber membranes than for flat sheet membranes due to the high packing density of hollow fiber modules. In addition, it tends to be lower at high influent flow rate (or flux) since scouring air flow does not increase proportionally to the influent flow. SADp also varies by the wastewater characteristics, SRT, water temperature, MLSS, foaming, etc, since these factors affect mixed liquor quality that in turn determines the sustainable permeate flow rate at a given scouring air flow.
Although comparing SADp of different membranes is hard, a general trend can be found in Table 1 and Fig. 1, where hollow fibers have lower SADp. In addition, as can be seen in the table, actual SADp observations in fields are higher than the guidelines suggested by manufacturers based on pilot tests under ideal flow conditions.
Table 1. Commercial membranes used in MBR
|Manufacturer||Configuration (Model)||Flux, net (m/d)||Area|
|Air, net (m3/hr)||SADm|
(m3 air/ m2membrane/hr)
(m3 air/ m3permeate)
|Remark||Raw data source|
|GE Water||HF(ZW500d)||0.6||98||18||0.18||7.3||Manufacture’s spec||Côté, 2004a|
|Econity||HF (4005CF)||0.4||1000||150||0.15||9.0||Personal communication|
|Koch Puron||0.14-0.53||Herold, 2011|
|Kubota||FS (EK200)||0.6||160||96||0.60||24||Pilot test||Adham, 2004|
|Siemens||HF(B10R)||0.58||37||13.4||0.36||16||Pilot test||Adham, 2004|
|Mitsubishi||HF(SADF)||0.72||29||13.7||0.47||16||Pilot test||Wei, 2006|
|A3 Water Solutions||HF(SADF)||0.6||140||28||0.2||8||Pilot test||Grélot, 2010|
|Asahi||HF (MUNC-620A)||0.6||25||6||0.24||9.6||Pilot test||Personal communication|
|GE Water||HF(ZW500a)||0.6||5,280||2,300||0.44||17||Rödingen, Germany||Brepols, 2004|
|HF(ZW500c)||0.6||4,088||1,577||0.39||15||Key Colony, FL||Survey|
|0.48||0.26||13||Pilot test||Côté, 2004b|
|?||?||?||?||15||Varsseld, Netherlands||Brepols, 2004|
Note: Design parameters were used for the calculation. Actual SAD can increase during dry season since flux decreases more sharply than aeration rate.
Though there are claims that the SADp of newly designed flat sheet cassettes is comparable to that of hollow fiber cassettes (Grélot, 2010), it has been widely observed around the world that hollow fibers have advantages over flat sheets in scouring air utilization by a factor of 1.5-2.5. In the meantime, MBRs with flat sheet membranes tend to require less biology air since some portion of oxygen supplied by scouring air can be utilized biologically.
According to the field survey performed by Barillon et al. (2013), aeration energy demands are equal or less for membrane scouring than for biological aeration in hollow fiber membranes (Fig. 2a). On the contrary, aeration energy demands are nearly double the energy demands for biological aeration in flat sheet membranes (Fig. 2b).
Fig. 1. SADp as a function of net flux in various MBR plants (Judd, 2007)
a) Flat sheet (30,000 P.E., 9.8 g/L MLSS, 47% annual average organic loading) b) Hollow fiber (5,000 P.E., 8.1 g/L MLSS, 51% annual average organic loading)
Fig. 2. Distribution of energy consumption of MBR units in biological steps (Barillon et al. 2013)
© Seong Hoon Yoon