Evaluation of Passive Nitrogen Removal Systems

Daniel P. Smith1

ABSTRACT

Removal of nitrogen from onsite wastewater using “passive” systems has been identified as a significant need by the Florida Department of Health (FDOH). The Florida Passive Nitrogen Removal Study was initiated with the objectives of performing a literature review of passive nitrogen removal systems, identifying conceivable media, elucidating the processes underlying media function, and performing experiments to support recommendations for onsite wastewater treatment applications. This paper describes the experimental approach and results of bench scale testing of passive nitrogen removal media treating septic tank effluent (STE). The operative definition of the term “passive” specified that systems incorporate reactive media, have no aeration pumps, and employ at most a single effluent dosing pump. These constraints required that gravity flow be one facet of system design and influenced design of the bench scale systems. A central modus of bench scale evaluations was a two-stage system of porous media filters, with an initial unsaturated media filter (ammonification and nitrification) followed in series by a saturated anoxic filter with reactive media (denitrification). Parallel two-stage systems contained unsaturated media columns followed in series by saturated media columns. The unsaturated Stage 1 columns received STE and included synthetic and natural media, and sorbing and reactive media. Stage 2 (saturated) columns received Stage 1 effluent and included synthetic and natural media, sorbing media, and reactive media to supply electron donor and alkalinity. Factors that were evaluated include influent and effluent water quality, dosing regime, applied loading rates (hydraulic, organic and nitrogen), and the Total Nitrogen:alkalinity ratio of influent. Keywords Onsite wastewater Nitrogen reduction Filtration Reactive media 1Applied Environmental Technology, Thonotosassa, Florida 33592

PASSIVE NITROGEN REMOVAL There are currently 2.6 million onsite wastewater treatment and disposal systems in the State of Florida, with over fifty thousand new connections in 2006/2007 (FDoH 2008). Technologies that increase nitrogen removal have been identified as one component of a strategy to reduce the potential effects of onsite nitrogen loading to sensitive environments (Roeder, 2007). The Florida Passive Nitrogen Removal Study was established to identify and evaluate passive treatment systems that are capable of achieving greater nitrogen reductions than exhibited by conventional septic tank/drainfield configurations. The Florida Department of Health (FDOH) provided a definition of “passive” as “A type of onsite sewage treatment and disposal system that excludes the use of aerator pumps and includes no more than one effluent dosing pump in mechanical and moving parts and uses a

1

reactive media to assist in nitrogen removal.” The specific interest is in approaches to onsite treatment that employ filter media, including reactive media, systems that do not employ aerators, and systems that minimize the need for liquid pumping. In the first task of the Florida Passive Nitrogen Removal Study was a literature review of passive technologies with the potential to enhance nitrogen removal from on-site wastewater treatment systems (Smith and Otis, 2007). The report proposed a coupled two stage filter system for passive removal of total nitrogen from septic tank effluent. The first stage is an unsaturated media filter for ammonification and nitrification. Pumping to the unsaturated filter will allow pressure and timed dosing. The second stage is a saturated anoxic filter with reactive media (denitrification). The two stage configuration is mandated by the obligatory biochemical sequence of aerobic nitrification followed by anoxic denitrification. The use of an unsaturated trickle flow media filter for initial nitrification is necessary because of the constraint that aeration pumps can not be used in the passive system. Flow from Stage 1 to 2 would be by gravity and pumping would not be required. The system would be deployed between the septic tank and the soil treatment/dispersal system of new or existing facilities. Nitrogen in septic tank effluent would be substantially removed before wastewater was directed to the soil for treatment or dispersal. The second phase of the Florida Passive Nitrogen Removal Study Task 2 is an experimental evaluation of candidate filter media and systems for total nitrogen removal from STE. Small scale testing will be performed to identify candidate media for subsequent evaluation using full scale onsite wastewater treatment systems. To perform the media evaluations, it was desired to conduct studies in a manner that closely resembles the functioning of an actual onsite system. The actual candidate media will be used, placed in appropriate depth and distribution. Dosed filter operation is preferable, similar to what could be established in an operating system.

FILTER MEDIA

Aerobic (Unsaturated) Filters Media properties significantly affect the performance of unsaturated aerobic filters. Smaller particle size distribution increases the specific surface area available for biofilm attachment and particle retention. Media in the upper filter layer will have a greater need to retain and biodegrade wastewater solids and larger particle size may beneneficial. Water retention capacity is a significant property of unsaturated filter media. High water retention will maximize the length of time that newly applied wastewater remains in the filter and increase contact time with media surfaces and attached microorganisms. Unsaturated media filters are four phase systems: solid media, attached microbial film, trickling wastewater, and the gas phase. The air filled porosity (i.e. gas phase) is that portion of the external porosity that is available for the transport of oxygen to surfaces throughout the media. High air filled porosity is desirable to maintain aerobic biochemical reactions reactions, particularly since higher hydraulic loading rates increase the pore water content and will possibly inhibit nitrification. The characteristics of the specific filter media interact with the design and operation of the unsaturated filter. The average hydraulic loading rate per filter surface area affects the average loading rates of organics and nitrogen. While the rate of wastewater generation is typically highly variable, filtration will be most effective when wastewater is

2

applied continuously or as small frequently applied doses. In an unsaturated filter, a small hydraulic application rate per dose will result in improved trickle flow contact of new wastewater parcels with surfaces and attached biofilms. If the hydraulic application rate is a small fraction of the total water retained within the unsaturated filter, breakthrough of constituents into filter effluent will be minimized. Anoxic (Saturated) Filters Anoxic saturated media filters form a second stage in the passive nitrogen removal system. Anoxic filters contain a “reactive” media that provide a slowly dissolving source of electron donor for reduction of nitrate and nitrite by microbial denitrification. Denitrifying microorganisms grow predominantly attached to the media surfaces. Water flows by advection through the media pores, where oxidized nitrogen species are consumed by attached microorganisms. Water saturation of the pores prevents ingress of oxygen, which could interfere with nitrate reduction. Factors influencing the performance of anoxic denitrification filters include hydraulic and nitrogen loading rates, media particle size and surface area, pore size, flow characteristics within the reactor, and length or depth of filter bed. Dissolution of reactive electron donor media must be sufficiently rapid to supply electron equivalents for nitrate/nitrite reduction and possibly other reactions. On the other hand, too rapid a dissolution rate would reduce the longevity of the media and could release higher concentrations of excess dissolution products in the effluent. Another factor influencing denitrification performance is the accumulation of suspended solids within the column, which could result in the development of preferential flow paths, decreased contact time of wastewater with media surfaces, and performance deterioration. Nitrate and nitrite removal over extended operation would be more likely if the anoxic filter received a first stage unsaturated filter effluent with low levels of suspended solids and biodegradable organics (BOD). The aspect ratio of the denitrification filter and flow entrance and withdrawal would also affect flow patterns and potential short circuiting. The effects of flow channeling on performance deterioration could possibly require maintenance or media replacement at time scales appreciably shorter than theoretical longevities based on stoichiometric requirements for denitrification. Media Media that will be evaluated are listed in Table 1 and shown in Figures 2 through 7. Zeo-Pure is a clinoptilolite with high water retention characteristics and a total cation exchange capacity for ammonium of 1.8 to 2.0 meq/g (Zeox Corp. 2008). Zeolite media have been shown previously to be highly effective in improving ammonium filtration under both steady and non-steady loadings (Smith et al. 2004). Livelite is an expanded clay which also has high water retention and porosity. Recycled rubber materials are produced by shredding of used tires and particle size reduction, and are available as tire chips and crumb rubber in particle sizes of 5 mm and less. Tire chips have been evaluated for use as a drainfield media, where they would function in an unsaturated filtration mode (Grimes et al. 2003). Elemental sulfur will be used as electron donor media is in the Stage 2, which will establish an autotrophic denitrification process in the anoxic filter. Elemental sulfur has been evaluated as denitrification filter media for onsite wastewater (Sengupta et al. 2006), groundwater (Darbi et al. 2003), aquaculture (van Rijn et al. 2006), and stormwater (Smith, 2008). Crushed oyster shell will be used as an alkalinity source, as sulfur-based autotrophic denitrification is an

3

alkalinity consuming biochemical reaction. Expanded shale is also included in the Stage 2 media mix. Utelite has an anion exchange capacity of 120 meq/100 g (Zhu et al. 1997), which would bind nitrate ions and possibly enhance performance resiliency under non-steady operation.

4

Table 1 Filter Media

MaterialBulk density,

lb/ft3Particle Size

Range Supplier

Zeo-Pure AMZ 8/20 Clinoptilolite 55 0.8 - 2.3 mm Ash Meadows, Armagosa, NV

Livlite Expanded Clay 41 3 to 5 mm Big River, Alpharetta, GA

Tire Crumb 25 0.3 - 5 mm Global Tire Recycling, Wildwood, FL

Elemental sulfur 77 2 - 4 mm Georgia Sulfur, Valdosta, GA

Oyster shell 82 3 - 15 mm Harold's Supply, Dover, FL

ACT-MX ESF-450 Utelite 54 0.4 - 4.5 mm ES Filter, Ogden, UT

EXPERIMENTAL SYSTEMS A schematic of the experimental filter columns is shown in Figure 1. Three filter systems will be evaluated, each consisting of an unsaturated filter followed by a saturated filter. Filters will be fabricated from 3 in. diameter tubing (unsaturated filters) and 1.5 in. diameter tubing (saturated filters), using a 1/8 inch screening for media support and retention. Filter columns will be constructed of materials with high contact angles for water sorption to minimize wall effects. The surface area of the filter media will also be twenty to fifty times that of wall area. Septic tank effluent will be applied to the surface of the first stage media, resulting in a downward percolation of wastewater over and through the media filter bed. The unsaturated pore spaces in the first stage media will allow air to reach microorganisms attached to the media surfaces, enabling aerobic biochemical reactions to occur. Effluent from the bottom of the first stage filter will be passed through a saturated anoxic horizontal flow filter that contains reactive media that supplies electron donor for denitrification. Following startup, the column systems will be operated for two months and monitored for nitrogen species and other water quality parameters. Of particular interest are the concentrations of nitrate, nitrite and total nitrogen in the second stage effluent. The configuration of the three 2 Stage filters is listed Table 3. The Stage 1 columns will use clinoptilolite, expanded clay, and tire crumb, with total media depth will of 24 in. Media stratification based on particle size is based on the expected progression of biochemical reactions within the filter media. The processes in the upper media layer include adsorption of wastewater particulates and colloids, hydrolysis and release of soluble organics, aerobic utilization of soluble organics, and biomass synthesis. In this region, the biochemical processing of organic matter between doses must keep up with the newly applied wastewater constituents from each dose. The greatest accumulation of organic and inorganic mass is expected in the upper layer, and the use of larger particle size media will provide greater space for accumulation of solids. Stratified media should enhance to potential for long term

5

operation while maintaining treatment efficiency. The use of finer particle sizes in lower depths will provide greater surface area for microbial attachment and a finer media for physical filtration, the later which could improve removal of pathogens and other wastewater constituents. The progression of coarser to finer media size through the filter will also enable coarser media to filter out larger particulates and protect the finer media that follows.

Septic Tank

Effluent

Peristaltic Pump

Support Screen

Stage 1 Media

24 in.

Flow Distributor

Stage 1 Effluent

Stage 2 Media

Stage 2 Effluent

Figure 1 Experimental Filter System Schematic Three Stage 2 columns will be constructed or unstratified media containing elemental sulfur, crushed oyster shell, and expanded shale (Table 2) of 24 in. media depth. Each filter will contain a 3:1 ratio of elemental sulfur to crushed oyster shell (vol./vol.), which has previously been shown to provide adequate alkalinity. The difference in the Stage 2 media composition is the fraction of expanded shale, which ranges from 0 to 40%. Expanded shale contains anion exchange capacity which can bind nitrate ions, potentially enhancing removal. In addition, higher expanded shale fractions are accompanied by lower fraction of elemental sulfur. The later could reduce the total surface area of elemental sulfur and the overall sulfur oxidation rate, resulting in less complete denitrification. Effluent sulfate levels could be reduced with lower sulfur fractions. The use of three sulfur fractions will allow this issue to be examined.

6

Table 2 Configuration of Two Stage Filters

Stage Filter Column ID, inch.

Total depth, inch

Media placement Media

1A8 in. clinoptilolite (2.3-4.8 mm) 8 in. clinoptilolite (0.8-2.3 mm) 8 in. clinoptilolite (0.5-1.1 mm)

1B8 in. expanded clay (3-5 mm) 8 in. expanded clay (0.8-2.3 mm) 8 in. expanded clay (0.5-1.1 mm)

1C8 in. tire crumb (3-5 mm) 8 in. tire crumb (1-3 mm) 8 in. tire crumb (0.4-1 mm)

2A 75% elemental sulfur 25% oyster shell

2B60% elemental sulfur 20% oyster shell 20% expanded shale

2C45 % elemental sulfur 15% oyster shell 40% expanded shale

Stratified

1.5 24.0 Nonstratified

Stage 1

Stage 2

3.0 24.0

The Stage 1 filters will be vertically oriented and Stage 2 filters placed horizontally (Figure 1). The Stage 1 filters will be supplied with septic tank effluent by a multi-head peristaltic pump, with a repeat cycle timer for dosing of once per one half hour (48 doses/day). The water elevation in the tube below the Stage 1 filter will provide hydraulic head for passive movement of water through the Stage 2 filter. A valve and sample port (with another valve) will be located in the tube below the Stage 1 filter. In normal filter operation, the sample port valve will be closed and the valve leading to Stage 2 will be open. The design of the filter system minimizes internal volumes within the connecting piping. Operation will be conducted at a hydraulic loading to the Stage 1 filters of 3 gal./ft2-day. Operating characteristics of Stage 1 and Stage 2 filters are shown in Tables 3 and 4. At 48 doses per day, a single dose will add a volume that is 6% of the water retained within the Stage 1 filter bed. The estimated average water residence time in the Stage 1 filter is 9 hr. (Table 3). An average water residence time of 12 hr. is provided in the Stage 2 filter (Table 4).

7

Table 3 Operating Characteristics of Stage 1 Filters (unsaturated)

Diameter, inch 3.0

Media depth, inch 24

Flow, gpd/ft2 3.0

Doses/day 48

Empty bed volume, liter 2.8

Resident water volume, liter1 0.21

Single dose volume / resident water volume 0.06

Average residence time, hour 9.0

1Assumes 50% external porosity, 15% water filled. Table 4 Operating Characteristics of Stage 2 Filters (saturated)

Diameter, inch 1.5

Media depth, inch 24

Flow, gpd/ft2 12.0

Empty bed volume, liter 0.69

Pore volume, liter1 0.28

Average residence time, hour 12.0

1Assumes 40% external porosity. Operation of columns was initiated on 12/20/2007. The systems will be operated for four weeks before monitoring is initiated. Target parameters are total nitrogen, ammonia, nitrate and nitrite, and other water quality parameters. Acknowledgements The author gratefully acknowledges the Florida Department of Health (FDoH) for providing funding for this project and the NOWRA Education Committee for reviewing this article.

8

REFERENCES

1. Darbi, A., T. Viraraghavan, R. Butler, and D. Corkal. 2003. Pilot-Scale Evaluation of Select Nitrate Removal Technologies. Journal of Environmental Science and Health Part A—Toxic/Hazardous Substances & Environmental Engineering, A38(9): 1703-1715.

2. Florida Department of Health 2008. Onsite Sewage Treatment and Disposal Systems Installed in Florida. Onsite Sewage Programs Statistical Data 1971 to 2007 http://www.doh.state.fl.us/environment/ostds/statistics/newInstallations.pdf

3. Grimes, B., S. Steinbeck, and A. Amoozegar. 2003. Analysis of Tire Chips as a Substitute for Stone Aggregate in Nitrification Trenches of Onsite Septic Systems. Small Flows Quarterly 4,4,18-23.

4. Roeder, E. 2007. A Range of Cost-Effective Strategies for Reducing Nitrogen Contributions from Onsite Sewage Treatment and Disposal Systems Task 4 of the 2006/2007 Wekiva Study, Bureau of Onsite Sewage Programs, Florida Department of Health, Tallahassee, FL http://www.doh.state.fl.us/environment/ostds/wekiva/task4/FinalReport.pdf

5. Sengupta, S. S. Ergas. S. Sengupta, E. Lopez-Luna, , A. Sahu, and K. Palaniswamy. 2006. Autotrophic Biological Denitrification with Elemental Sulfur or Hydrogen for Complete Removal of Nitrate-Nitrogen from a Septic System Wastewater. Water, Air, and Soil Pollution: Focus, 6, 111-126

6. Smith, D. and R. Otis. 2007 Florida Passive Nitrogen Removal Study Literature Review and Database Prepared For Florida Department of Health, Tallahassee, FL November 12, 2007.

7. Smith, D. 2008. Sorptive Media Biofiltration for Inorganic Nitrogen Removal from Stormwater, J. Irrigation and Drainage Engineering, American Society of Civil Engineers, in press.

8. Smith, D., M. Flint, and J. Merriam. 2004. Zeolite Filters: An Innovative BMP for Enhanced Nitrogen Removal from Stormwater, Proceedings of Water Environment Federation Annual Convention (WEFTEC), New Orleans, Louisiana.

9. Van Rijn, J., Y. Tal, and H. Schreier. 2006. Denitrification in Recirculating Systems: Theory and Applications. Aquacultural Engineering, 34,6,6, 364-376.

10. Zeox Corporation. 2009. Zeo-Pure, Ash Meadows, NV.

11. Zhu, T., P. Jenssen, T. Maehlum, and T. Krogstad. 1997. Phosphorus Sorption and Chemical Characteristics of Lightweight aggregates (LWA) - Potential Filter Media in Treatment Wetlands. Water Sci. Technol., 35, 5, 103-108.

9

Figure 2 Stage 1 media: Ash Meadows clinoptilolite (0.5 to 1 mm)

Figure 3 Stage 1 media: Livelite expanded clay (0.5 to 1 mm)

10

Figure 4 Stage 1 media: Rubber crumb (0.5 to 1 mm)

Figure 5 Stage 2 media: Elemental sulfur pastille (0.5 to 1 mm)

11

12

Figure 6 Stage 2 media: Utelite expanded shale (0.5 to 1 mm)

Figure 7 Stage 2 media: Crushed oyster shell (0.5 to 1 mm)