MBR technology proves beneficial for plants of different sizes
In Canton, Ohio, a few weeks after ringing in 2014, four contractors submitted bids to update an aging extended aeration wastewater treatment plant (WWTP). Based on bid numbers, it will cost about $80 million to upgrade the 42-million-gal-per-day (mgd) plant to a membrane bioreactor (MBR), or less than $2 per gallon of water treated.
On a much smaller scale, in 2013, a failing trickling filter was converted to novel single-stage MBR in Pembroke, Mass. The high-rate MBR installed in Pembroke—about 1,000 times smaller than the proposed system in Canton—has reliably produced reuse quality effluent containing less than 7 mg/L of total nitrogen since August 2013 and has become the model for other projects in the area looking for better treatment. These two plants are extreme examples—one large and one small—of a rapidly growing trend in wastewater treatment: Using MBR technology to rehabilitate older conventional plants is not only good for the environment; it is good for saving money.
The Large: Canton, Ohio
In building what will be the world’s largest MBR, the city of Canton is carrying on a tradition dating back to 1893, when it built the first WWTP in the state of Ohio. The main driver for upgrading the existing activated sludge plant is a new phosphorus limit of 1 mg/L. Other factors that played a role in the decision-making process included new air quality permits related to incinerators, pending nitrogen limits of less than 10 mg/L, constructability concerns and total cost of ownership.
In addition to MBR technology, five other solutions were considered as viable options to meet the new phosphorus limit, including simple coagulant addition and integrated fixed-film activated sludge (IFAS). Initially, coagulant addition seemed like a workable option, but increased sludge production and high chemical costs proved prohibitive. In fact, the estimated increase in solids handling costs due to added chemicals was a major concern, given that part of the overall project included demolition of existing incinerators. In the end, the MBR option was 28% more cost-effective to build and 23% less expensive to own over time, considering total operating costs. In addition to lowering costs, turning the old activated sludge plant into MBR also:
The existing plant is not designed for nutrient limits and is hydraulically unable to handle short-term peak flows during storm events. For most of the year, influent flow rates range from 29 to 32 mgd, but on rare occasions, transient flows have reached 108 mgd. The new MBR plant is being designed to handle an average annual flow of 39 mgd and a maximum monthly flow of 42 mgd with provisions to manage storm flow. The design also calls for operation at elevated mixed liquor suspended solids (MLSS) concentrations so that the biological process can fully nitrify and partially denitrify treated wastewater.
The ability to run at higher MLSS concentrations is one reason flat-plate technology was selected for this project. At the same time, given the fixed footprint, the ability to deliver enough oxygen to the process in fairly shallow tanks (approximately 12 ft) also was an important consideration. Ultimately, the membrane equipment selected must operate efficiently at an MLSS concentration of nearly 13,000 mg/L while transferring oxygen at a rate of 1% per foot of submergence as part of air scouring.
The type of membrane used was only one of many important decisions in the process of design optimization. For example, there are numerous ways to make the plant work hydraulically and ensure water will flow properly from point A to point B during staged commissioning and once the project is complete. There also were a variety of approaches to making the biological process work. Not all of the possibilities considered were equal, however, and it is well established that hydraulics and process conditions are interrelated in a submerged MBR—one impacts the other and vice versa.
Leveraging more than 200 process simulations and using data from a full-scale pilot, the project team devised a design that essentially converts each of the process trains into smaller independent WWTPs. Each of these 7-mgd WWTPs will comprise an anaerobic zone, an anoxic zone, a pre-aeration zone, and an MBR zone (for filtration and nitrification). Factors considered during design included:
From an overall standpoint, and to more readily compare other projects, the scope of work for Canton includes:
Invariably, questions will come up regarding the applicability of the approach to other smaller plants that may not see the same economies of scale.
The Medium: SBR Retrofit Projects
The vast majority of publicly owned WWTPs permitted in the U.S. are much smaller than the Canton plant, ranging in size from 1 to 5 mgd. Moreover, according to U.S. Environmental Protection Agency data, roughly 450 of the nearly 16,500 permitted plants are sequencing batch reactors (SBRs), but that number could be much higher. Quotes from three representative mid-size plants seem to be consistent with Canton estimates regarding total cost of ownership and how the MBR stacks up against conventional options.
“The electrical is higher, the solids production is lower and we are now able to use non-potable water in the plant; we couldn’t with SBR. With these three factors, it’s about the same cost for the MBR and SBR,” said Sue Lawrence, supervisor of a 0.9-mgd MBR.
Therefore, for very large and mid-size plants data suggest MBR upgrades may provide workable solutions for activated sludge plants and SBRs. But what about small, decentralized systems? Does MBR make sense there, too? Also, can other types of conventional plants be converted into MBRs?
The Small: Pembroke, Mass.
The advantages of MBR technology are scalable from a plant the size of Canton’s down to small, decentralized (point-of-use) facilities. In fact, the main advantage of MBR—the ability to increase process treatment capacity in a given volume—can be further magnified beyond what is envisioned for Canton by a factor of almost three using a “high-rate” MBR operating at MLSS concentrations up to 30,000 mg/L.
A WWTP permitted under the Massachusetts groundwater discharge standards of treatment was commissioned last year in Pembroke, Mass. With a rated capacity of 40,000 gpd, this single-stage MBR once was a failing trickling filter plant that had been in operation for almost five years before replacement. The old plant had been unable to meet total nitrogen (TN) limits, resulting in an Administrative Order on Consent to upgrade the plant. Challenged by a limited budget, the decision was made to implement an MBR solution.
The total cost to retrofit the Pembroke WWTP was just more than $900,000. Given the size of the job, the normalized cost per gallon is much more than for the Canton plant, but upgrading to MBR was still significantly cheaper and faster than other options. In terms of delivery, the facility was converted in just four months by maximizing the use of existing infrastructure. In operation now for more than a year, the plant has been consistently providing effluent quality of biochemical oxygen demand less than 3 mg/L; total suspended solids less than 2 mg/L and TN less than 7 mg/L—well below the permit requirements.
The New MBR Reality
At a total installed cost of less than $2 per gal, the Canton WWTP upgrade project may prove to be a harbinger of a new approach to big plant upgrades. However, the approach is not limited to big plants. From small to large, MBR technology is being used to effectively upgrade older conventional plants for a variety of reasons, including lowest total cost of ownership.
Dennis Livingston is director, MBR systems, for Ovivo. Livingston can be reached at [email protected] or 512.834.6019.