The peak hydraulic capacity of submerged membrane bioreactor (sMBR) systems is generally on the order of twice the rated average capacity of the treatment plant. In certain regions where heavy rainfall and/or rapid snow melt are prevalent, short-term hydraulic peaks can vary between two to 10 times the average plant flow. These peak flow periods can be as short as a few hours, or as long as days and even weeks.
For systems with such large transient peaking factors, it is cost prohibitive to build and operate an sMBR. A cost-effective technology called STORMBLOX has been developed that can efficiently trim (or manage) short-term peak events in MBR plants while meeting the most stringent permit limits. Moreover, this new green technology does not require air scouring of any kind, significantly reducing energy usage and carbon footprint.
Figure 1: Peak flow management process
A peak flow management process called STORMBLOX was developed to operate in conjunction with sMBR systems as a means of augmenting treatment capacity during periods of high flow. As shown in Figure 1, the first step of the process is fine screening (FS), followed by ultrafiltration (UF) for removal of total suspended solids, pathogens (such as bacteria and viruses) and turbidity. Filtrate can subsequently be passed through activated carbon (AC) for removal of biochemical oxygen demand (BOD) and zeolite media for removal of ammonia (as ammonium ion). Retentate (rejected solids) is periodically sent to the sMBR process where the solids are eventually removed via waste activated sludge (WAS).
STORMBLOX systems are designed to produce permit-compliant effluent using direct filtration of raw wastewater followed by contact with well established, naturally occurring media. All of these operations are widely used in the industry, but the success of this technology depended on the proper pre-treatment of influent combined with the right membrane technology. iSEP UF technology (manufactured by the TriSep Corp.) proved well suited for this innovative treatment method.
iSEP Technology
Direct low-pressure filtration of raw wastewater will typically lead to rapid fouling and unsustainable performance. However, during STORMBLOX testing it was discovered that pre-conditioning wastewater with aluminum sulfate (alum) significantly reduced the rate of fouling when using iSep 500 UF membranes. Further testing indicated that membrane performance actually improved without air scouring.
Figure 2: iSep Module
iSep membranes utilize a 0.03-micron proprietary hydrophilic membrane chemistry specially designed for difficult waste streams with high fouling potential. iSep is an integrated submerged UF element where the membrane and tank have been integrated into a single module (Figure 2). Negative pressure is applied to generate a vacuum and “pull” water through the membrane. In typical applications, air is bubbled up through the flow channels to actively scour the membrane surface and remove particulate matter.
However, integrated as a part of a STORMBLOX system, no air is required and retentate containing the alum dosed for fouling control is discharged to the main sMBR system. Once added to the mixed liquor in the biological process, the alum will remove some phosphorus through sweep flocculation and other mechanisms. Additionally, alum is known to improve the filterability of mixed liquor. Eventually, added alum is wasted along with biological solids.
Design
Extensive testing indicates net fluxes in excess of 30 gal per sq ft per day are sustainable with typical backwashing and alum dosing. Trials were conducted using standard municipal strength waste and diluted waste streams more representative of storm flows. Under all conditions energy usage for full treatment was on the order of 0.10 kWh/cu meter or about a quarter of the most efficient sMBR . Similarly, the footprint to handle storm flows is roughly a quarter that of an sMBR system. Capital and operating expenses proved economical and are a strong function of peaking factors, duration of peak events and sustainable flux, as sensitivity analysis will show.
