NITRIFICATION OF A LAGOON
EFFLUENT USING FIXED FILM MEDIA:
Aerated lagoon systems for wastewater treatment are commonly used in rural states where land area is plentiful and population densities low. They are intentionally designed to be simple in operation and to require a minimal degree of operator attention. These systems are capable of achieving conventional secondary treatment with little operational flexibility. In recent years, due to water quality concerns, a number of lagoon systems in New Hampshire have been assigned year round ammonia limitations in their NPDES discharge permits. More are expected to follow. In order to meet an ammonia removal requirement, the degree of treatment needs to increase considerably. Conventional engineering practice is to convert these systems to nitrifying activated sludge. While guaranteed to meet strict year round ammonia limits, conversion to activated sludge does have substantial drawbacks; namely high costs and increased complexity. This paper will present the results of a two-year, small-scale pilot project designed to evaluate the potential of using fixed film media to nitrify a lagoon effluent.
BACKGROUND & THEORY
For many years it has been debated as to whether aerated lagoons are capable of achieving biological nitrification as currently designed. In order to nitrify, which is the biological conversion of ammonia to nitrate through the actions of specialized nitrifying bacteria, sufficient quantities of the bacteria must reside within the system. Nitrifying bacteria are considered attached growth organisms, meaning that they attach to the surface of an object as opposed to assuming a free-swimming mode. Otherwise, they may be lost to the effluent or deposited to the sludge blanket. In the activated sludge process, the floc particle serves as the object. In trickling filters and RBC抯, the media itself or the biological growth residing on the media surface serves as the attachment site.
The dilute liquid environment found within a lagoon does not provide many suitable attachment sites to sustain large populations of bacteria. The solids concentration within a lagoon cell, usually <100 mg/l, cannot compare to the mixed liquor concentrations of activated sludge, often >2000 mg/l. Nitrifying bacteria, however, may be found in limited quantities on side slopes, baffles, or the sludge blanket/water interface.
Several studies undertaken in New Hampshire in recent years have determined that lagoons can and do nitrify to significant degrees, often achieving < 1.0 mg/l ammonia nitrogen in the effluent. However, it appears to be a highly seasonal occurrence, more prevalent in the summer and early fall than at other times of the year. The degree of nitrification can also vary significantly from year to year, sometimes going from a nitrifying mode one year to a non-nitrifying mode the next. It is also a plant specific occurrence; some systems may nitrify completely while others only partially or not at all.
Temperature has the greatest impact. The rate of nitrification is maximized when water temperatures are > 20oC, a condition only found during the summer and early fall. The rate of nitrification ceases or slows down considerably as temperatures approach 5oC. Beyond this, little to no nitrification is thought to occur. In northern climates such as New Hampshire, lagoons develop an ice cover and temperatures drop to 0 degrees C.
When all of these factors are considered together, it is very difficult or nearly impossible for lagoon based systems to meet year round ammonia limits consistently without substantial modifications. Permit limits in New Hampshire vary depending upon the sensitivity of the receiving stream, but currently range anywhere from 0.81 ? 6.3 mg/l during the summer season and 2.38 ?16.3 mg/l during the winter season.
It has been estimated by others that converting a moderately sized lagoon system of < 1.0 MGD design flow to activated sludge could cost anywhere from $3,000,000 to $5,000,000. O&M costs also increase due to solids handling processes, expanded process control, higher electrical consumption, and maintaining the multitude of mechanical equipment associated with activated sludge.
One potential alternative to activated sludge that has seen limited full-scale application in this country is the incorporation of fixed film media within an existing lagoon system in order to remove ammonia. The media, placed on the bottom of one or more cells of the lagoon, provides ample surface area attachment sites to support large masses of nitrifiers. This has been attempted in Colorado and Ohio with some apparent success. If this technology proves feasible in the long run, it could retain the simplicity of the lagoon system, keep operating costs down, and meet ammonia limits at the same time.
PILOT STUDY RATIONALE AND GOALS
Knowing the harsh winter climate of New Hampshire, it would be difficult to convince any community facing ammonia limits to invest in a full-scale application of a technology that has not been proven. It was decided that further investigation was warranted and that it should be done on a pilot-scale basis. There are numerous types of fixed film media systems now on the market and several were evaluated for use in this study. The selected system was a product called BioMatrix Looped Cord Media, marketed at the time by BioMatrix Technologies of Lincoln, Rhode Island. Looped cord media has its beginnings in Japan and is constructed of braided linear composite threads. It has the appearance of a very thin nylon rope with all of the individual strands pulled out to form small loops along its length. This is similar in construction to the types used in Colorado and Ohio. The BioMatrix system is designed specifically for wastewater treatment.
There are also many other potential products manufactured today that are not designed for use in wastewater treatment that may work as well as those materials specifically designed for this purpose. Alternative materials could ultimately prove to be less expensive than traditional fixed film systems if they can be modified for use in this application. The pilot-study was then designed to be a side-by-side comparison of two different types of media; one designed for wastewater treatment and one that has had no prior application in wastewater treatment. The alternative media needed to meet certain criteria: #1. It must provide adequate surface area with sufficient roughness in order to allow a biological growth to grab hold. #2. It must be non-toxic. #3. It must not biodegrade in water. Coarse scrubbing pad material manufactured by the 3M Company of St. Paul, MN was chosen to be the alternative media. This material is the same type of abrasive fabric found on the bottom of floor scrubbing machines and on the back of some kitchen scrubbing sponges.
The goals of the pilot study were the following:
The pilot study was designed to operate for a minimum of two years. This was done to evaluate performance over different seasons and temperature ranges.
PILOT STUDY LOCATION
The Town of Exeter, NH WWTF was selected as the site for the pilot study. Exeter has a three cell aerated lagoon system with a 3.0 MGD design flow. Their current ammonia as N limits are 0.81 mg/l as a monthly average from May ?October and 2.38 mg/l from November ? April. The facility discharges to the tidal Squamscott River on New Hampshire抯 seacoast. The outfall pipe is exposed during low tide. Because of the exposed outfall pipe the dilution factor is calculated as zero. Exeter has one of the strictest ammonia limits in the state because of the zero dilution. Historically, effluent ammonia concentrations range from <1.0 mg/l in the summer and fall to >20 mg/l during winter and early spring. Clearly, the facility cannot achieve compliance on a consistent basis. Exeter was selected to host the pilot study because of their willingness to participate, they have ammonia limits, and they were actively engaged in design work with a consulting firm for possible plant upgrades.
PILOT STUDY DESCRIPTION
The pilot study consists of a side-by-side comparison of the BioMatrix fixed film system and the 3M scrubbing pad material. A modular storage tank with a capacity of 4,000 gallons and having dimensions of 19 feet long by 7 feet 9 inches wide and 4 feet deep was used to contain the pilot project. A dividing wall was constructed down the length of the tank to separate it into two totally isolated tanks. The Biomatrix system occupied one tank and the 3M fabric occupied the other tank.
3800 linear feet of the BioMatrix media was purchased and divided equally amongst four supporting frames having dimensions of 3?X 3?X 3? The frames with attached media arrived fully assembled. The 3M material was purchased in a roll 132 feet long by 3 feet wide. Four supporting frames with similar dimensions were manufactured on-site using PVC pipe. The 3M fabric was cut into 3 foot squares and equal amounts were supported on each PVC frame.
Each tank was separated into 5 equally sized compartments through the use of cross baffling. One cube of media occupied each of the first 4 compartments; BioMatrix placed on one side and 3M on the other side. A fine bubble diffused aeration system commonly used in lagoons provided the necessary aeration and mixing. The final compartment of each tank was used to provide a quiescent settling zone for sloughed solids. This compartment was unaerated and did not contain any media. Effluent flowed from these final compartments on a continuous basis.
The tanks were continuously fed with fully treated (BOD < 30 mg/l) effluent from the final lagoon cell. A submersible pump located in a sampling manhole (prior to chlorination) was used to feed the first bay of each fixed film system with lagoon effluent. The flow was equally split as much as possible between the two tanks. It would then proceed in a plug flow fashion through all four compartments before finally exiting compartment #5.
Each tank had an approximate volume of 1872 gallons with 6 inches of freeboard. The influent flow to each tank can be varied somewhat from 0.5 to 7.0 gpm. This provides a detention time ranging from 2.6 days per tank during winter conditions to 0.19 days per tank during summer conditions. It was anticipated that a longer detention time would be necessary during the winter.
The modular tank was situated outside and exposed to the elements during the spring, summer, and fall. For winter operation the tank was covered with an all-weather tent style vinyl shelter large enough to allow someone to walk inside and monitor tank conditions. The tank and associated piping were insulated to prevent freezing problems. No supplemental heat was added. Effluent from both fixed film systems was returned to the final lagoon.
Influent to the tanks (lagoon effluent) and the BioMatrix and 3M effluents were monitored weekly for ammonia nitrogen (NH4-N) and pH. Nitrite (NO2-N), nitrate (NO3-N), alkalinity, and ORP were monitored bi-weekly or more often as required. Temperature and flow rate measurements were done daily. NH4-N analysis was done on-site using a Hach DR2000 spectrophotometer or Hach DR700 colorimeter. A preserved sample for NH4-N was also collected at the same time for analysis by a certified lab. All NH4-N results discussed in this paper are certified lab results with the exception of several that are based on on-site analysis due to a lack of certified results for various reasons.
The pilot study became operational in mid October 1999. This paper discusses results from October 28, 1999 thru March 14, 2002 for a total of 125 weeks of operation.
The monthly average effluent ammonia concentrations can be found on the graph in Figure 1.
In general, the two systems combined removed an average of 41% of the applied ammonia. The 3M fabric performed slightly better overall, removing 45% vs. 36% for the BioMatrix system. Seasonal variations definitely occurred, with lowered performance during the winter months. Figure 2 shows a graph of the monthly averages for percentage of ammonia removed. The production of nitrates, as shown in Figure 3, confirmed that the ammonia reductions observed during the study were due to nitrification, and not some other method such as ammonia stripping. Some degree of nitrification was always maintained even during the coldest of temperatures. For purposes of discussion, the monthly average results from Figure 1 will now be broken down into seasonal events.
First Fall/Winter/Spring - This period represents the first 36 weeks of operation, from 10/28/99 - 6/29/00. During the first 12 weeks, 10/28/99 - 1/12/00, the weather was relatively mild. During this time the 3M system removed on average 90% of the applied ammonia and BioMatrix removed 82%. For a period of three weeks lasting from 12/28/99 - 1/12/00, at water temperatures averaging 2-3oC, the 3M system fully nitrified. Effluent ammonia concentrations were 0.1, 0.1, and 0.28 mg/l. The BioMatrix effluent was only slightly higher at 1.4, 0.7, and 1.89 mg/l. Lagoon effluent ammonia concentrations during this same period were 9.9, 12.0, and 13.6 mg/l.
A severe cold snap developed during late January 2000. Water temperatures fell to 0oC and remained at this temperature through the end of February. Effluent ammonia concentrations increased and percent removals decreased. It did appear, however, that nitrification was never completely lost as there was always some ammonia reduction.
For a 3 week period in February 2000, an effluent recycle system was operated within the 3M tank. This system consisted of a submersible pump that was suspended in the final 3M compartment. Treated effluent was recycled at a high rate back to the first 3M compartment. This was done to see if effluent recycling, by passing portions of the treated effluent back through the media multiple times, would improve performance in cold temperatures. Percent removals did increase from 31% for the week prior to the initiation of recycling, to 98% during the last week of recycling. 3M抯 effluent ammonia decreased from 9.5 mg/l to 0.42 mg/l during this period. Unfortunately, over the course of the three weeks, the submersible pump artificially raised the temperature of the 3M tank from 0 Degrees C to 4.6 Degrees C. Because of the unintentional temperature increase, it was impossible to determine whether recycling alone played a role in the improved performance. For comparison purposes, the BioMatrix effluent ammonia concentration for this same three week period, without the benefit of effluent recycle and at tank temperatures of 0 Degrees C to 1 Degrees C, decreased from 16.6 mg/l to 13.55 mg/l.
Complete nitrification (<0.5 mg/l effluent ammonia) at 8 Degrees C was obtained by both systems on March 30, 2000. This degree of treatment continued throughout the summer. The lagoon itself did not achieve complete nitrification until the beginning of July.
A lagoon turnover due to warming water temperatures was experienced from May 25, 2000 - June 29, 2000). This resulted in a month long increase of elevated lagoon effluent ammonia concentrations. The temporary increase was completely nitrified by both fixed film systems.
First Summer/Fall - This period encompasses the timeframe of July 2000 thru the first half of November 2000. The lagoons completely nitrified throughout this period. The low ammonia concentrations entering the pilot tanks resulted in little nitrification actually taking place within the fixed film systems and did not prove to be an adequate test under summer conditions.
Second Fall/Winter/Spring ?This period represents the last half of November 2000 ?mid May 2001. This winter proved to be much colder than the first and the onset of cold weather began considerably earlier. System performance during this period did not compare to that experienced during the first winter. Effluent ammonias from both fixed film systems tracked very closely with the lagoon effluent. Percent removals for January and February 2001 averaged only 4% for BioMatrix and 5.5% for 3M.
The effluent recycle experiment was repeated again. Each system was recycled for one month, BioMatrix during January and 3M during February. This time the results appear to indicate that effluent recycling did not have any significant effect on system performance.
For the second consecutive spring, the 3M media achieved complete nitrification very early in the season. Beginning on March 30, 2001, at a water temperature of 3.6oC, 3M discharged an effluent containing only 0.5 mg/l NH4-N, for a removal rate of 96%. The BioMatrix system during the same week discharged 9.03 mg/l and the lagoon discharged 12.8 mg/l.
Feed From the Second Lagoon - During the first summer and subsequent fall, July 2000 thru early November 2000, the lagoons nitrified completely. Ammonia concentrations exiting the final lagoon were consistently <0.5 mg/l, most often as low as 0.1 mg/l. Because of the low ammonia feed entering the pilot tanks, it was impossible to evaluate the potential of these fixed film systems to nitrify under warm temperature conditions. More importantly, however, it was not known what impact a low ammonia feed would have on the ability to sustain a large population of nitrifying bacteria on the media throughout the summer. It was feared that there could be a continuous die-off or their ability to reproduce could be compromised if their food source (ammonia) was seriously depleted for any substantial length of time. If reproduction and sustainability were indeed affected, it could have a negative impact on cold temperature nitrification if an adequate biomass has not been developed by the onset of cold temperatures.
Because of these concerns, it was decided that for the second summer of operation the feed to the pilot study would be switched from the final lagoon to the beginning of the second lagoon. This feed point was selected because of its proximity to the pilot tank. The theory behind this change was that the ammonia concentration at the beginning of the second lagoon, based on prior history, should consistently be higher throughout the summer than the ammonia found in the final effluent. If this proved to be true, then a healthy population of nitrifiers should be sustainable throughout the summer and would help to ensure that a good population be on hand as soon as cold temperatures approached. This may help to improve overall winter performance. Of course, it was also anticipated that the fixed film systems would be treating a higher BOD load from this lagoon.
Feeding from the second lagoon began on May 17, 2001 and ended on November 16, 2001, for a total of 27 weeks. Initially, lagoon ammonias were high and partial nitrification was occurring within the fixed films. Beginning in early July and lasting through mid October, the second lagoon also completely nitrified, once again limiting the ammonia feed to the pilot tanks. For the second summer in a row, the fixed film systems were unable to be fully tested under summer conditions. The feed point was reverted back to the final lagoon on Nov 20, 2001.
Third Winter/Spring ?This period, December 2001 ?mid March 2002, was very mild in terms of temperature and below average snowfall. The lagoons remained free of ice for much of the winter, despite 0oC water temperatures. Lagoon effluent ammonia concentrations peaked higher this February, averaging 23.7 mg/l, than at any other time during the study. Percent removal rates for this period averaged 30% for Biomatrix and 43% for the 3M fabric. The effluent recycle experiment was repeated again for 6 weeks on the 3M side only during January and the first half of February. During this period of recycle, the 3M fabric removed 48% of the applied ammonia. The removal rate for BioMatrix without the aid of recycle was 27%.
Other Observations ?During each fall large amounts of filamentous algae growth attached directly to the fixed film, sometimes forming dense floating mats on the water surface. It is not known if this interfered in any way with nitrifier attachment to the media. If nitrifiers did cling to the algae, then they were lost as soon as the algae sloughed from the media, which tended to happen during early winter. Biomass growth in general (excluding algae) was always heavier in the winter and always sloughed off in the summer.
Alkalinity did not appear to be a limiting factor during this study. Nitrification rates exceeding 90% removal within the two fixed film systems were often observed with alkalinity in the single numbers to low teens. The pH of the pilot tanks rarely dropped below 6.5 and most often was around a pH of 7.