PA/600/R-12/655 | September 2012 | www.epa.gov/gateway/science
   United States
   Environmental Protection
   Agency
Summary Report:
Pilot Study of an Innovative Biological Treatment
Process for the Removal of Ammonia from a
Small Drinking Water System
                              j

     Office of Research and Development

-------
                                                                 EPA/600/R-12/655
                                                                 September 2012
                            Summary Report:


Pilot Study of an Innovative Biological Treatment Process for the

    Removal of Ammonia from a Small Drinking Water System

                                 Prepared by

                                Darren A. Lytle
                       U.S. Environmental Protection Agency,
                        Office of Research and Development
                   National Risk Management Research Laboratory
                     Water Supply and Water Resources Division
                        Cincinnati, Ohio 45268, United States

                                 Colin White
                             University of Cincinnati
                        Cincinnati, Ohio 45221, United States

                                 Dan Williams
                       U.S. Environmental Protection Agency,
                        Office of Research and Development
                   National Risk Management Research Laboratory
                     Water Supply and Water Resources Division
                        Cincinnati, Ohio 45268, United States

                                 Lauren Koch
                             University of Cincinnati
                        Cincinnati, Ohio 45221, United States

                                Emily Nauman
                          Pegasus Technical Services, Inc.,
                        Cincinnati, Ohio 45219, United States
                               September 2012

-------
                                          Notice
 The U.S. Environmental Protection Agency, through its Office of Research and Development, funded and
managed, or partially funded and collaborated in, the research described herein. It has been subjected to
    the Agency's administrative review and has been approved for external publication. Any opinions
expressed in this paper are those of the author (s) and do not necessarily reflect the views of the Agency,
    therefore, no official endorsement should be inferred. Any mention of trade names or commercial
               products does not constitute endorsement or recommendation for use.

-------
        of





Notice	i



Table of Contents	ii



List of Figures	iii



List of Tables	iii



List of Appendices	iv



1.  Background	1



  1.1 Ammonia in Drinking Water Sources	1



  1.2 Community with Elevated Ammonia Levels	1



  1.3 Ammonia Treatment Options	2



2.   Biological Water Treatment Technology Pilot Study	5



  2.1 Collaboration	5



  2.2 Research Approach	5



  2.3 Pilot Technology Description	6



3.   Operations, Materials,  and Methods	8



  3.1 Pilot System Operation	8



  3.2 Water Quality Analysis	9



4.   Results of the Pilot Study	11



  4.1 Important Dates	11



  4.2 General Water Chemistry	11



  4.3 Removal of Ammonia in Source Water	11



  4.4 Removal of Iron from Source Water	23



  4.5 Other Water Quality Parameters	25



  4.6 Assessment of Bacterial Population Based on HPCs	27



5.   Discussion and Summary	28



  5.1 Discussion	28



  5.2 Summary of Key Findings	29



  5.3 Future Work/Questions	30



6.   References	31



APPENDIX A	32

-------
Figure 1. Map of ammonia levels in Iowa	3
Figure 2. Schematic of the pilot biological ammonia removal treatment technology system	6
Figure 3. Three gravels used in the contactors of the piloted biological ammonia removal treatment
technology system	7
Figure 4. Pilot biological water treatment system for ammonia removal at the Iowa study site	7
Figure 5. Nitrogen content of treated water from Contactor 1	13
Figure 6. Contactor 1 flowrate and hydraulic loading rate	14
Figure 7. Contactor 1 nitrogen mass balance between influent and effluent	15
Figure 8. Nitrogen content and DO of finished water from Contactor 1	17
Figure 9. Filter 1 data: A) Nitrogen content of finished water from Filter 1, B) Filter 1 flowrate and
loading rate, C)  Filter 1 nitrogen balance effluent, and D) Effluent ammonia and dissolved oxygen	18
Figure 10. Nitrogen content of finished water from Contactor 2	19
Figure 11. Contactor 2 flow and loading rate	20
Figure 12. Contactor 2 effluent ammonia and DO	21
Figure 13. Filter 2 data. A) Nitrogen content of finished  water from filter 2, B) Filter 2 flowrate and
hydraulic loading rate, and C) Filter 2 ammonia and oxygen	22
Figure 14. Iron in raw water and treated water through Contactor 1 and Filter 1	24
Figure 15. Iron in raw water and treated water through Contactor 2 and Filter 2	24
Figure 16. Raw water temperature and DO with average air temperature	25
Figure 17. Total  alkalinity of raw, Contactor 1 effluent and  Filter 1 effluent	26
Figure 18. Heterotrophic plate counts (HPCs) in raw, Contactor 1 effluent and Filter 1 effluent	26
Figure 19. Heterotrophic plate counts (HPCs) in raw, Contactor 2 effluent and Filter 2 effluent	27
Table 1. Source Water Quality	4
Table 2. Timeline of Operational Changes for Contactor 1 and Filter 1	9
Table 3. Water Quality Analyses Performed and Methods	10
Table 4. Water Quality Summary (Average ± Standard Deviation (n))	12
Table 5. Final Design and Operating Parameters	28

-------
A 1. Contactor 1 effluent temperature and DO	32



A 2. Contactor 2 effluent temperature and DO	33



A 3. Contactor 2 and Filter 2 effluent Temp and DO	33



A 4. Nitrogen content of raw water	34



A 5. Nitrogen content of finished water from Filter 3	34
                                                                                           IV

-------
1.   Background
1.1 Ammonia in Drinking Water Sources
 Many regions in the United States have excessive levels of ammonia in their drinking water sources
(e.g., ground and surface waters) as a result of naturally occurring processes, agricultural and urban
runoff, concentrated animal feeding operations, municipal wastewater treatment plants, and other
sources. Ammonia is not regulated by the U.S. Environmental Protection Agency (EPA) as a contaminant.
Based on a 2003 World Health Organization (WHO) assessment, ammonia levels in groundwater are
typically below 0.2 milligrams per liter (mg/L), and does not pose a direct health concern at levels
expected in drinking water (WHO 2003); however, it may pose a concern when  nitrification of significant
levels of ammonia from the source water occurs in the drinking water distribution system. Specifically,
this nitrification, which is the conversion of the ammonia to nitrite and nitrate by bacteria, leads to
water quality issues, such as potential corrosion problems, oxidant demand, taste and odor complaints,
and elevated nitrite levels (Bremer et al., 2001; Fleming et al., 2005; Lee et al., 1980; Odell et al., 1996;
Rittman & Snoeyink, 1984; Suffet et al., 1996).
Ammonia in water may also pose problems with water treatment effectiveness. For example, in source
waters containing both ammonia and arsenic, the ammonia may negatively impact the removal of
arsenic by creating a chlorine demand, therefore reducing the availability of chlorine needed to oxidize
the arsenic (Lytle et al., 2007). Lastly, water systems that have ammonia  in their source water and desire
to maintain a free chlorine residual will need to add additional chlorine to overcome the demand of
ammonia. Clearly, the complete oxidation of source water ammonia prior to or  as part of the water
treatment process would eliminate the potential negative impacts of nitrification on distribution system
water quality.

1.2 Community with Elevated Ammonia Levels
Many regions in the Midwest are particularly impacted by ammonia in their source waters from natural
geology, agricultural runoff, and other farming practices. For example, the State of Iowa has a
widespread distribution of ammonia in well waters across its communities (Figure 1). Water quality
testing of the source groundwater in one of the communities (population approximately 873) (Table 1)
showed that, on average, ammonia levels were 3.3 mg as nitrogen (N)/L. Although the focus  of this
report is on ammonia contamination, it is relevant to note that the water samples averaged 0.82 mg/L
of iron. Similar to ammonia, iron in drinking water does not pose a direct health concern. However,
there is an EPA recommended, non-enforceable National Secondary Drinking Water Regulation
Standard of 0.3 mg/L for iron, which is based on aesthetic and technical issues, rather than health-based
concerns. Specifically, iron in the water can cause a metallic taste, discoloration of the water, staining of
faucet and fixtures, and sediment build-up. Given the negative issues associated with high ammonia
levels in drinking  water, and with the added issues from the high levels of iron, there is a clear need to
establish effective treatment approaches to address these issues. Furthermore, the State of Iowa

-------
Department of Natural Resources (DNR) can request water systems to monitor nitrite and nitrate in
their distribution systems, should they suspect that nitrification is occurring in their distribution system.

1.3 Ammonia Treatment Options
The most commonly used water treatment options for addressing elevated ammonia in source waters
are the formations of monochloramine and breakpoint chlorination. Breakpoint chlorination results in
the removal of ammonia as nitrogen gas by a chemical reaction with chlorine; typically in the range of 8
to 11 times the mg N/L ammonia present. For a community with a water source such as the community
chose in this study, this would be a very high chlorine dose. The formation of monochloramine involves
the addition of chlorine to concentrations where ammonia is not removed but rather bound to chlorine.
Other approaches including ion exchange with zeolites, reverse osmosis (RO), advanced oxidation,  and
air stripping, are capable of removing ammonia from water, but are relatively complex, expensive,  or
have limited applications.
Although often performed unintentionally, biological ammonia "removal"1 is another treatment
approach to reduce source water ammonia. The process relies on bacteria to convert ammonia to
nitrate. As a result, a more biologically-stabile water is produced, nitrification in the distribution system
is not an issue, and free chlorine residual is easily achieved. Biological conversion of ammonia (NH3) to
nitrate (NO3~) involves a two-step sequence of reactions mediated by two different genera of bacteria:
Nitrosomonas and Nitrospira. These autotrophic bacteria derive energy for cellular functions from the
oxidation of ammonia and  nitrite, respectively. Nitrosomonas are responsible for the oxidation of
ammonia, in the form of ammonium (NH4+), to nitrite (NO2~) according to the reaction:
                              NH4+  +  1.5 02 -^ N02- + H20 + 2H+                  (1)

 Nitrospira subsequently oxidizes nitrite to nitrate, as follows:

                                     NO2" + 0.5 O2 ->  NO3"                        (2)

By summing these equations, the overall nitrification reaction is obtained:

                               NH4+ +  2O2 -> NO3"  +  2 H+ + H2O                  (3)

It should be noted that these equations are net reactions involving a complex series of enzyme-
catalyzed intermediate steps. Nitrification produces free protons, H+ which readily consume available
bicarbonate ions (HCO3~), thereby reducing the buffering capacity of the water. In addition, nitrifying
bacteria consume CO2 to build new cells. The total consumption of alkalinity by nitrification is 7.1 mg as
CaCO3 per mg NH4+- N oxidized (US EPA, 1975). The oxygen demand of nitrification is also significant. For
complete nitrification, 4.6 mg O2 is required per mg NH4+- N oxidized (US EPA, 1975; US EPA 1993).
Other factors that affect nitrification include phosphate concentration, pH, and water temperature. All
organisms including nitrifying bacteria require phosphorus  to build cell mass, with approximately 3% of
1 The terms "removal" and "oxidation" will be used interchangeably throughout this document. We use "removal"
to represent the conversion of ammonia to nitrate and/or nitrite by biological oxidation. We recognize that
treatment does not physically remove ammonia-nitrogen but rather converts the form of nitrogen (i.e., total of
ammonia, nitrite, and nitrate).

-------
dry weight consisting of phosphorus. Microorganisms use phosphate as the source of phosphorus for
the synthesis of structural and physiological biomolecules such as deoxyribonucleic acid (DNA),
phospholipids (membranes), teichoic acid (cell walls), and most importantly, as inorganic phosphorus in
adenosine triphosphate (ATP) synthesis. Without ATP, the cellular metabolism (i.e. nitrification) cannot
proceed and the cells either become dormant or die. Some organisms are more sensitive to phosphate
starvation than others, and in the case of nitrification, ammonia oxidizing bacteria are less sensitive than
nitrite oxidizing bacteria (de Vet et a/., 2012; Scherrenberg et a/., 2011; Scherrenberg et a/., 2012).
Numerous laboratory studies have cited the optimum pH for complete nitrification is between 7.4 and
8.0; although in practice, the  bulk water pH may deviate from this value while nitrification remains high
(Shammas, 1986). Temperature can impact growth rate and metabolism by slowing or destroying
necessary enzymes and proteins involved in physiological processes. Laboratory studies have
demonstrated that the growth rate of nitrifying bacteria to be negatively impacted by temperatures
below 10°C, although adjustments to the treatment process can be made to enhance nitrification in
colder climates (Andersson, et a/., 2001).
                                                        J5^°~
Ammonia Levels  [O Q
O 0-.5 mg/L        U
O -5-1.0 mg/L
O 1.0-3.0 mg/L
• 3.0-5.0 mg/L      ( Q(O
0 5.0-10.0 mg/L      )
Figure 1.
         Map of ammonia levels in

-------
Table 1. Source Water Quality
Alkalinity
Fe
Mn
P
TOC
Zn
S
CI"
Mg
NH4
NO2
NO3
pH
Temp °C
358mgCaCO3/L
0.82 mg/L
0.01 mg/L
0.07 mg PO4/L
1.06 mg/L
0.26 mg/L
33 mg/L
<5 mg/L
33 mg/L
3.3 mg-N/L
0.04mg-N/L
0.02 mg-N/L
7.4
15.8

-------
2.   Biological Water Treatment Technology Pilot Study
2.1 Collaboration
The small community study site in Iowa does not have a centralized water treatment or a drinking water
distribution system. Following extensive flooding to the region in 2008, support to build the necessary
infrastructure to supply the community with potable drinking water was put into place. The treatment
system needed to be designed to address elevated levels of both iron and ammonia in the source water.
The State of Iowa's DNR requested assistance from EPA's  Office of Research and Development (ORD) to
develop an appropriate treatment system to address the source water concerns. Specifically, ORD's
experience in applying biological water treatment to remove ammonia from water was requested. As a
result, the State of Iowa DNR, and EPA ORD and Region 7, conducted a pilot study to evaluate the
impact of biological water treatment on ammonia oxidation.
Specifically, the pilot is based on an EPA-patented approach (Figure 2) to address elevated levels of
ammonia as well as iron in the source water (Patent No. US 8, 029,674). The treatment system relies on
bacteria for the conversion of ammonia to nitrate; provided the raw ammonia levels are lower than the
nitrate MCL of 10 mg N/L, the approach can be effective and relatively simple. The pilot system was
designed and built by EPA staff, and installed in March 2011 (Figure 4). In a collaborative effort, EPA and
pilot site staff coordinated system operation and maintenance, as well as water sample collection and
analysis.

2.2 Research Approach
Nitrification is a two-step, microbiological process that requires oxygen (aerobic) to oxidize NH4to NO2,
and then to NO3. The entire process requires approximately 4.5 mg of O2/mg of NH4-N in the source
water.  Because the groundwater in the study community has low oxygen (3.6 mg O2/L) and elevated
ammonia of 3.3 mg N/L as well as reduced iron of 0.82 mg/L (Table 1) that also exerts an oxygen
demand, more than 13.5 mg O2/L would be necessary to address the demand due to the ammonia (and
iron). Aeration is a necessary feature of the biological ammonia treatment system; however, the
traditional configuration of aeration followed by filtration (e.g., iron removal) including biologically-
active filtration is not sufficient to address the oxygen demand to meet the treatment objectives of the
community's water system.
The amount of  oxygen that can be added to the water is controlled by the saturation limit of oxygen in
water, which in most drinking waters including the study community's, is well below the total oxygen
requirements of treatment. The  EPA's experience with microbiological systems that do not provide
sufficient oxygen to a nitrifying system has shown that the result is incomplete nitrification or the
production of elevated nitrite levels in the finished water. Given the drinking water standard for nitrite is
only 1 mg N/L, concerns for potential exceedances exist where source water ammonia levels are greater
than 1 mg N/L.  Therefore, an innovative approach to introducing oxygen to the treatment system in the
small community was necessary  to meet the treatment objectives. Aerating with pure oxygen could
provide super saturated oxygen conditions and sufficient oxygen, however there are safety issues
associated with flammable gases and filter binding associated with gas bubbles can also be an issue.

-------
2.3 Pilot Technology Description
The ammonia biological removal treatment pilot system evaluated is based on an EPA patented design
(US 8,029,674 B2 awarded on 10/2/2011) seen in Figure 2. The pilot consisted of three pairs of 3-inch
(7.62 cm) diameter columns in series built from clear PVC and other common plumbing materials (Figure
2). Each pair consisted of one column or "contactor" filled with 30 inches (76.2 cm) of gravel (Figure 3) in
series with a second column or "filter" filled with anthracite (20 inches [50.8 cm] deep) over sand (10
inches [25.4 cm] deep); each contactor contained a different size gravel :  % inch (6.35 mm), 1/£ inch (12.7
mm), and 1 inch (25.4 mm) (Figure 2). The contactors were aerated from  the bottom, such that air
bubbles flow upward countercurrent to the water flow (downflow) using  a diffuser connected to a gas
pump at a rate of 2.5 L/min (0.66 gpm).
In this configuration, the water in the contactor was always saturated with respect to dissolved oxygen
throughout the gravel media bed despite  the demand from nitrification process and iron oxidation. The
gravel in the contactor was solely to serve as a growth support for nitrifying bacteria where nitrification
occurs. Gravel allowed bacteria attachment and growth yet eliminated the potential for "clogging" of
the media and regular backwashing, and allowed air bubbles to move through the contactor. Oxidation
of ferrous iron in the source water also occurs in the contactor, but no iron removal should occur.
Various flowrates were considered during pilot evaluations. The filter was intended to remove  iron
particles and potentially bacteria, and can also provide biological oxidation of excess ammonia and/or
nitrite that exit the contactor as a  result of incomplete nitrification. With  regards to the latter, the filter
serves as a polishing step and safeguard against disruption in operation of the contactor which could
result, for example, in excess nitrite formation. Effluent water from the filter is routed to a clear well,
that when full, can be used to backwash the filters, or overflow to the sanitary sewer.
 Water from the source
  enters the column
    from the top
                                           Downflow
      Gravel-filled column
        = Contactor
        (Labeled C1,
          C2,&C3)
       Each contactor
      contains a different
        size of media
                t
                Air
              bubbles I PUMP
                                                          Anthracite over sand
                                                               = Filter
                                                          (Labeled F1.F2.&F3)
                                                                        Water leaves pump at base of
                                                                        filter and goes to clear well
              1                 2                 3

Figure 2. Schematic of the pilot biological ammonia removal treatment technology system.

-------
               Small gravel
                    Cl
Medium gravel
        C2
Large gravel
      C3
                                            50 mm
Figure 3. Three gravels used in the contactors of the piloted biological ammonia removal treatment technology system
         Figure 4. Pilot biological water treatment system for ammonia removal at the Iowa study site.

-------
3.   Operations,  Materials,  and Methods
3.1 Pilot System Operation
The pilot system (Figure 4) was operated on a continuous basis (24 hours per day, 7 days per week) with
the exception of a few instances where pumps were replaced or other maintenance actions occurred
(<24 hours downtime or the system was intentionally shut down to evaluate the impact of doing so) for
over extended periods of time. Raw water from the small community's existing well and drinking water
was not chlorinated or treated in any way prior to supplying the pilot system. Treated water and excess
filter backwash water was routed to the on-site sanitary sewer.
Field operating and water quality measurements were collected by the city and included headless,
flowrates, temperature, dissolved oxygen, and pH. Dissolved oxygen, pH, and temperature were
measured using an HQ40d meter with an  LD101 dissolved oxygen  probe and PHC281 pH probe (Hach
Company, Loveland, CO). Filters were backwashed using filter effluent water on a weekly basis.
Backwashing was achieved by expanding the bed by 50% for 15  minutes. Contactors were backwashed
on a case by case basis using raw water. Contactor gravel did not expand during backwashing. A total
volume of 12.5 gallons (47.3 L) was used to backwash the contactor for approximately 5 minutes at rate
of 2.5 gallon/min (gpm) (9.45 L/min).
A number of parameters were varied and modifications to the pilot system operation were made to
optimize nitrification; these included changes to loading rate, media surface area, and a chemical feed
addition. Changes to pilot system operation, water quality, and other notable condition changes are
summarized in Table 2. Filter loading rate changes were made by adjusting the flowrate through the
pilot columns by adjusting the pump speed. For example, contactors began the study with a loading rate
of 4.0 gpm/ft2 (9.76 m/hr) and ended the  study at 2.0 gpm/ft2 (4.88 m/hr). Filters averaged 1.5 gpm/ ft2
(3.66 m/hr) over the duration of the study.
Due to limitations inherent in a pilot-scale system, the contactor oxygen diffusers required field
maintenance. The maintenance interval was determined based on field measurements of effluent
dissolved oxygen.
At 190 days into the study, a phosphate chemical feed was added  only to Contactor/Filter 1. The target
orthophosphate concentration was 0.4 mg PO4/L Orthophosphate was provided by the EPA in the form
of technical grade Na3PO4-12H2O (Fisher Scientific) suspended in deionized water. This solution was
added to 20 L of raw water in a carboy and injected into Contactor 1 at 2 mL/min via a peristaltic pump
To assess the effect of doubling the contactor depth, Contactor  1 effluent was routed to Contactor 2 and
served as the sole influent  at 360 days into the pilot. The effluent of Contactor 2 was routed to filter 1
for polishing. By routing Contactor 1 (30 inches [76.2 cm] of  media) to Contactor 2 (30 inches [76.2 cm]
of media), effectively doubled the contactor bed depth to 60 inches (152.4 cm). Contactor 3 was not
providing new data and was shut down on January 20, 2012.

-------
          Table 2. Timeline of Operational Changes for Contactor 1 and Filter 1

               Date        ET, days                Description of Change
3/28/2011
4/26/2011
5/6/2011
8/22/2011
9/9/2011
9/21/2011
9/29/2011
11/8/2011
12/20/2011
1/17/2012
2/21/2012
2/22/2012
2/27/2012
3/6/2012
4/24/2012
24
53
63
171
190
201
209
249
292
319
354
355
360
376
417
Backwash-Contactor 1
Backwash-Contactor 1
CI2 backflow event
Flows recorded at top of column
PO4feed (6 g)-Contactor I/Filter 1
Contactor 1 pump failed (<24 hours)
New operator
Flow change-Contactor I/Filter 1
Backwash-Contactor 1
Aerator blowout-Contactor 1
Aerator blowout-Contactor 1
Backwash-Contactor 1
Bed depth increase-Contactor 1 & 2
Flow change-Contactor I/Filter 1
PO4feed(3g)
3.2 Water Quality Analysis
Community staff collected weekly water quality samples, while making routine measurements and
shipped them on ice overnight to the US EPA Office of Research and Development (ORD) in Cincinnati
for analysis. Water samples were collected from the raw water and effluent of all contactors and filters.
The following water samples were collected on a weekly basis:
    •   250 ml for inorganic analysis
    •   60 ml for metals analysis
    •   250 ml for bacteriological analysis
    •   40 ml for organic carbon analysis

Upon arriving to EPA, the samples along with the chain of custody, were removed from the cooler,
preserved accordingly, and submitted for analysis. Ammonia, nitrite, and nitrate analysis were typically
performed on the same day the cooler arrived (approximately 24 hours after sampling). All water
analyses were performed according to EPA or Standard  Methods (Table 3).

-------
          Table 3. Water Quality Analyses Performed and Methods
                 Analysis
Method
Method #
Reference
Total Alkalinity
Ammonia (as N)
Chloride
Nitrate & Nitrite
(asN)
Orthophosphate
As, Pb, U, Se, Bi
Al, As, Ba, Be, Bi,
Ca, Cd, Cr, Cu, Fe,
K, Mg, Mn, Na, Ni,
P, Pb, S, Sb, Sulfate,
Si, Silica, Sn, Zn
TOC
Temperature
Total Coliforms*
E. coli.*
HPC
Potentiometric
Titration
Automated
Colorimetric
Potentiometric
Titration
Automated
Colorimetric
Automated
Colorimetric
ICP-MS
ICP-AES
Combustion
Thermocouple
Culture
Culture
Culture
2320 B.4.6
350.1
4500-CI D
353.2
365.1
200.8
200.7
5310 C
17.1
9223B
9223B
9215C
Std. Methods1
EPA Methods2
Std. Methods1
EPA Methods2
EPA Methods2
EPA Methods3
EPA Methods3
Std. Methods1
EPA Methods1
Std. Methods1
Std. Methods1
Std. Methods1
*lndicates random sampling

1 Standard Methods for the Examination of Water and Wastewater," 18th Edition (1992).
2 USEPA, "Methods for Chemical Analysis of Water and Wastes," EPA-600/14-79-020 (1983).
3 USEPA, "Methods for the Determination of Metals in Environmental Samples," EPA-600/14-91-010 (1994).
                                                                                               10

-------
4.    Results of the Pilot Study
4.1 Important Dates
There are a number of operating changes and other events that occurred over the course of the pilot
study that are worth noting because they had a direct impact on the results and proceeding discussions.
Events including changes in contactor/filter flowrates (loading rates), the addition of phosphate feed,
backwash events, aerator clean-outs, and operator changes have been documented (listed in Table 2)
and will be referred to when appropriate.

4.2 General Water Chemistry
Extensive water quality analysis of the site's source water, as well as the pilot contactor and filter
effluents over the entire pilot study, is summarized in Table 4. The source water was a relatively hard,
high alkalinity groundwater with calcium and magnesium levels averaging 79 and 33 mg/L, respectively,
or a total hardness  of 332 mg CaCO3/L, a total alkalinity of 357 mg CaCO3/L and a pH of 7.1. Iron levels
averaged 0.63 mg/L although the concentration varied as indicated by the relatively large standard
deviation, and ammonia averaged 3.2 mg N/L. Sulfate, chloride, and silica averaged 94 mg SO4/L, 5 mg/L
and 7.1 mg SiO2 /L, respectively. Orthophosphate was very low, averaging 0.032 mg PO4/L, and total
phosphorus was at  the detection limit of 0.005 mg P/L. Manganese, nitrite, and nitrate were at or near
the respective method detection limits; strontium averaged 1.1 mg/L; and TOC averaged 1.3 mg C/L.

4.3 Removal of Ammonia in Source Water
Contactor I/Filter 1. Contactor 1 was operated for approximately 55 days before nitrite levels started to
increase suggesting the initiation of the biological nitrification process within the contactor (Figure 5).
The increase was brief, however, and nitrite only reached 0.2 mg N/L before the level suddenly dropped
back to non-detectable (<0.01 mg N/L) (Figure 5). It was discovered that at 63 days, the pilot system
inadvertently received chlorinated water as a result of inadequate or failed backflow prevention
measures ahead of  the pilot system, the extent and time-frame to which was uncertain. The presence of
chlorine presumably halted biological activity in the contactor and , as a result, was attributed to the
decrease of nitrite generation. Proper backflow prevention was installed immediately after the event.
Following plumbing modifications to address backflow issues, nitrite levels  began to increase again by 95
days (Figure 5). Nitrite steadily increased to a peak of 0.4 mg N/L by 140 days. A similar decrease in
ammonia through the contactor over the same time period was observed. No nitrate was produced in
the contactor during this time.
                                                                                           11

-------
Table 4. Water Quality Summary (Average ± Standard Deviation (n))
Analyte
Ba
Ca
Cl
Fe
K
Mg
Mn
Na
NH3
NO2
NO3
o-PO4
P
SiO2
Sr
SO4
Total
Alkalinity
Total Nitrogen
TOC
Zn
PH
DO
HPC
Temperature
Detection Limit (mg/L)
0.001
0.01
5
0.001
0.3
0.005
0.001
0.03
0.03 (mg-N/L)
0.01 (mg-N/L)
0.02 (mg-N/L)
0.025 (mg PO4/L)
0.005 (mg P/L)
0.02 (mg SiO2)
0.001
0.003 (mgSO4/L)
1 (mg-CaC03/L)
0.01 (mg-N/L)
0.1 (mg-C/L)
0.0005
0.1
0.01 (mg-02/L)
1 (CFU/mL)
o.rc
Raw
0.028±0.001(45)
78.71±2.96(45)
5±1(52)
0.63±0.489(45)
5.2±0.1(45)
33.01±1.247(45)
0.008±0.004(45)
32.54±1.78(45)
3.24±0.35(52)
0.05±0.04(53)
0.03±0.02(53)
0.032±0.015(56)
0.013±0.01(45)
7.07±1.11(46)
1.114±0.053(45)
94.07±14.52(46)
357±2(46)
2.86±0.49(36)
1.3±0.7(42)
0.2193±0.2077(45)
7.1±0.32(52)
3.64±1.08(49)
10,631(37)
14.8±2.5(42)
Contactor 1
0.029±0.002(44)
78.6±3.09(44)
5±1(52)
1.523±1.916(44)
5.1±0.2(44)
32.987±1.306(44)
0.009±0.004(44)
32.51±1.46(44)
1.63±1.26(52)
0.52±0.68(53)
1.19±1. 17(53)
0.026±0.004(23)*
0.082±0.086(33)**
DL(24)
0.117±0.019(20)*
7.32±0.37(44)
1.114±0.055(44)
95.86±2.98(44)
344±10(47)
2.94±0.46(35)
1.3±0.4(40)
0.4045±0.3976(44)
7.36±0.42(51)
8.25±2. 13(47)
54,763(36)
14.8±2.2(41)
Filter 1
0.028±0.002(45)
78.08±2.83(45)
5±1(51)
0.019±0.016(45)
5.1±0.2(45)
32.948±1.274(45)
0.006±0.004(45)
32.5±1.42(45)
1.14±1.31(52)
0.39±0.76(53)
1.79±1.54(53)
0.027±0.006(23)*
0.053±0.036(33)**
DL(24)
0.021±0.008(21)*
7.18±0.32(45)
1.109±0.053(45)
95.64±2.89(45)
342±9(46)
3.1±0.45(37)
1.5±1.1(42)
0.1564±0.1803(45)
7.52±0.26(52)
8.02±1.93(47)
39,914(37)
15.4±2.8(42)
Contactor 2
0.03±0.004(44)
79.09±3.31(44)
5±1(51)
3.267±4.78(44)
5.2±0.2(44)
33.234±1.333(44)
0.01±0.007(44)
32.62±1.58(44)
1.91±1.05(51)
0.94±0.83(52)
0.53±1.07(52)
0.037±0.031(55)
0.032±0.109(44)
7.49±0.56(44)
1.12±0.054(44)
96.44±3.1(44)
343±10(46)
2.92±0.46(36)
1.3±0.6(41)
0.5739±0.607(44)
7.46±0.19(51)
9.23±1.31(46)
50,380(36)
14.7±2.3(40)
Filter 2
0.028±0.002(43)
78.06±3.8(43)
5±1(45)
0.054±0.17(43)
5.1±0.3(43)
33.04±1.232(43)
0.007±0.008(43)
32.35±1.04(43)
1.45±1.28(45)
1.73±2. 16(45)
0.53±0.73(45)
0.031±0.017(45)
0.016±0.02(43)
7.2±0.29(43)
1.112±0.055(43)
96.07±3(43)
343±13(41)
3.09±0.48(36)
1.3±0.7(41)
0.1654±0.1966(43)
7.54±0.17(43)
8.3±1.11(42)
36,436(26)
15.7±3.1(35)
Contactor 3
0.034±0.02(42)
79.03±3.99(42)
5±1(42)
8.681±24.39(42)
5.2±0.8(42)
33.152±1.731(42)
0.009±0.006(42)
32.63±2.2(42)
2.08±1.01(42)
1.23±1.06(42)
0.1±0.31(42)
0.03±0.012(42)
0.038±0.147(42)
7.76±1.47(43)
1.128±0.059(42)
96.45±4.17(43)
345±10(38)
2.95±0.43(36)
1.3±0.6(42)
0.7336±1.1754(42)

-
62,204(26)
15.1±2.7(32)
Filter 3
0.028±0.003(39)
78.88±2.96(45)
5±1(52)
0.031±0.489(45)
5.2±0.1(45)
33.135±1.247(45)
0.007±0.004(45)
32.65±1.78(45)
1.48±0.35(52)
1.59±0.04(53)
0.34±0.02(53)
0.028±0.015(56)
0.011±0.009(39)
7.18±0.37(39)
1.12±0.053(45)
96.73±3.31(39)
345±2(46)
3.05±0.49(36)
1.2±0.7(42)
0.1187±0.2077(45)
-
-
31,360(26)
16.3±3.4(29)
*Before phosphate feed
"After phosphate feed
                                                                                                   12

-------
                    WJ
                    O
                                100
                                         200
                                                  300
                                                           400
                                                                    500
                                     Elapsed Time (days)


A

	
	
NO-,
NO3
NH4 Raw
PO4 addition
Diffiisei blow-out
Bed dqjth increased
Diffiiser blow-out
                         Figure 5. Nitrogen content of treated water from Contactor 1.
The progression of bacterial acclimation and nitrification within the contactor was incomplete, and
unexpectedly and unacceptably slow. Considering variables that could impact the nitrification process,
the relatively high initial filter loading rate (flowrate) of 4.3 gpm/ft2 (10.5 m/hr) (Figure 6) through the
contactor was thought to potentially be related. The loading rate was incrementally decreased to 3
gpm/ft2 (7.3 m/hr) then 1 gpm/ft2 (2.4 m/hr) (Figure 6) between 140 and 160 days. The decrease in
loading rate resulted in an immediate increase in nitrite production and an equivalent ammonia
decrease to nearly 1 mg N/L. Still, no nitrate was produced in the contactor during this time. Although
some improvement was observed (i.e., more ammonia was oxidized), the progress was still very slow,
nitrite levels leaving the contactor approached the 1 mg N/L MCL and no signs of further oxidation to
nitrate were observed.
                                                                                             13

-------
c
I
s
01
                    o
                    E
                       1000
                        800
                        600
                        400
                        200
                                                             Flow
                                                             Loading Rate
                                                                        3
                                                        •3

                                                        J
                           0        100       200       300       400      500
                                        Elapsed Time (days)

                           Figure 6. Contactor 1 flowrate and hydraulic loading rate.
Previous work (Lytle et al., 2007) indicated the complete oxidation of ammonia to nitrate, or complete
acclimation of bacteria after start-up of a new biologically active nitrifying filter could take as little as 70
days. In addition, nitrite release generally occurs as a relatively short spike that falls off rather quickly as
nitrite oxidizing biofilm establish and nitrate is generated. Iowa's water in this study, however, had
nearly 3 times the ammonia concentration as the 2007 work. Other factors that impact biological
nitrification such as nutrient requirements were considered.
Phosphorus is an important nutrient and necessary physiological component of bacteria. The source
water contained very little "natural" orthophosphate (0.03 mg PO4/L) (Table 3). Insufficient phosphorus
that is necessary for cell physiology was considered as a possible issue, and therefore it was elected to
add phosphate ahead  of the contactor. At 190 days, orthophosphate was added to Contactor 1's feed
water at an arbitrary target dose of 0.4 mg PO4/L Measurable orthophosphate levels following chemical
feed only increased by 0.05 mg PO4/L (Table 3). Total phosphorus, however, increased by 0.1 mg P/L (0.3
mg PO4/L), which indicated that phosphate was likely bound to iron particulates or other solids in the
system.
Ammonia  in Contactor 1 effluent dropped below 0.05 mg N/L immediately following phosphate addition
(Figure 5). Nitrite initially spiked to nearly 3 mg N/L immediately after phosphate addition and then
rapidly dropped to 0.6 mg N/L within 30 days. During the same period of time, nitrate increased to as
high as 2.5 mg N/L but then decreased to and stabilized at approximately 1.5 mg N/L by 220 days.  The
rapid progression of nitrite to nitrate was almost immediately halted between 220 and 240 days, at
which time ammonia levels increased back to 1.4 mg N/L.
                                                                                             14

-------
At approximately 240 day, the contactor loading rate was increased to 2.5 gpm/ft2(6.1 m/hr) which did
not appear to impact nitrogen balances. Between 240 and 290 days, still no improvement in ammonia
oxidation was observed. During this time, loading rate was gradually dropped to 1.6 gpm/ft2(3.9 m/hr)
but no obvious improvement in ammonia oxidation was noted. Unfortunately, although phosphate
stimulated nitrite oxidation and enhanced ammonia oxidation, it was still not complete (Figures 5 and
7). Nitrite levels remained at or near the MCL for nitrite, where the goal was to completely oxidize
ammonia to nitrate in the contactor. At day 292, the contactor was backwashed for the first time not
because of headless build-up, but rather as a maintenance step and to remove some of the build-up of
iron and biomass on the gravel. After backwashing, nitrite levels dropped to near detection limit and
nitrate levels increased by a similar amount, but ammonia levels remained largely unchanged.
Oxygen is also a key parameter in the nitrification process, whereby 4.6 mg O2/L is necessary to oxidize 1
mg N/L ammonia to nitrate. Further, there is also a connection between oxygen levels and kinetic
requirements associated with molecular diffusion. Close examination of oxygen levels in the contactor
effluent showed that the increase in ammonia at approximately  220 days was directly linked to a sudden
decrease in oxygen in the contactor from approximately 8 mg O2/L down to 5.5 mg O2/L,  presumably
because of the onset of improved nitrification (Figure 8).
                                    100        200        300
                                         Elapsed Time (days)
                                                                  400
                                 NO;
                                 NO3
                                • Total RinvNitrosen
                   Figure 7. Contactor 1 nitrogen mass balance between influent and effluent.
                                                                                           15

-------
Once the oxygen drop issue was recognized, the system operator "blew-out" the air diffuserto remove
any biofilm and/or deposit that could have blocked the air diffuser. Immediately after cleaning the
diffuser, ammonia, and nitrite decreased and oxygen and nitrate increased (Figure 5 and 7). Dissolved
oxygen levels dropped slowly again shortly after normal operation resumed (Figure 8). A more
aggressive aerator blow-out was performed and further improvements were immediately realized. By
the second blow-out (~350 days), ammonia levels remained low or non-detectable, and the system
operated ideally. Providing adequate oxygen (to near  saturated oxygen levels) through the contactor, as
well as orthophosphate, were enough for the contactor alone to achieve the desired ammonia reduction
at a typical filter loading rate to many iron removal plants.
The phosphate feed was shut-off twice toward the end of the study (>350 days) after optimized
ammonia oxidation was realized to simulate chemical feed failure events. During the first event, the
feed was only turned  off for 24  hours prior to sampling. No degredation of ammonia oxidation was
noted nor was nitrite  generated. Later, the feed was discontinued for three straight weeks, during which
time, no degredation  of ammonia oxidation nor nitrite formation was noted. The results suggest
orthophosphate accumulated in the contactor (likely bound to iron particles) and was still biologically-
available.
The primary intent of the dual media filter that followed the contactors was to remove iron particles
that developed in the contactors. The filters were also biologically-active and provided protection by
oxidizing excess ammonia and nitrite that passed through the contactor. Ammonia, nitrite, and nitrate
levels entering Filter 1 were those exiting Contactor 1  (Figure 5 and 7). Ammonia oxidation to nitrite in
the filter began at about 95 days (same as contactor) and increased steadily to produce a very
concerning 2.8 mg N/L nitrite by 190 days (Figure 9a). As with the contactor, no nitrate was formed. The
filter loading rate during this time was between 1 gpm/ft2 (2.4 m/hr) and 1.5 gpm/ft2 (3.7 m/hr) up to
160 days (Figure 9b). The loading rate was reduced for a brief period to 0.5 gpm/ft2 (1.2 m/hr), which
had no obvious impact on ammonia oxidation. The addition of phosphate at 190 days caused an
immediate decrease in the filter effluent nitrite level to near detection and immediate increase of
nitrate to 3 mg N/L (ammonia was near detection limit). Ammonia levels increased occasionally but
remained <0.6 mg N/L for the period between 220 days and 319 days, and nitrite levels remained very
low during this period (Figure 9c). This time period corresponded to the dissolved oxygen issues in the
contactor which appeared to carry over to incomplete oxidation of ammonia in the filter. Dissolved
oxygen levels in the filter were near 6 mg/L during this time (Figure 9d). Once the oxygen concentration
was re-established following the diffuser blow-out, ammonia levels remained near the detection limit.
The filter was operated at a loading  rate of 2 gpm/ft2 (4.9 m/hr)  by the end of the study.
Clearly, the filter improved overall water quality by polishing contactor effluent. Most notably,
comparison  between  Figures 7 and 9c indicate the degree to which nitrate formation was enhanced in
the filter effluent (more green shared area).
                                                                                           16

-------
                  100
    200         300        400

Elapsed Time (days)
500
                            NH4
                            DO
                            Bed depth increased
                            PO4 addition
                            Diffiisei blow-out
                            Difiuser blow-out
Figure 8. Nitrogen content and DO of finished water from Contactor 1.
                                                                                     17

-------
                                                              -100
                                                              300
                                                           OS
                                                            *   0
                                                              100
                                                                                                                I
                      Elapsed Time (days)
                                                            100      :00      300      -100      500

                                                                 Elapsed Time (days)
    100       200        .400       400

1 NH4     Elapsed Time (days)
I NO,
|NOj
- Total RitwNitiooai
                                                                              Elapsed Time (days)
                                                                            NH4
                                                                  	PO4 addition
                                                                  	Diffusci blow-out
                                                                  	  Piifusei blow-out
Figure9. Filter 1 data: A) Nitrogen content of finished water from Filter 1, B) Filter 1 flowrate and loading rate, C)
                Filter 1 nitrogen balance effluent, and D) Effluent ammonia  and dissolved oxygen.
                                                                                                                   18

-------
Contactor 2/Filter2. Contactor 2 was operated near identically to Contactor 1 with the exception that
orthophosphate was not added at any time and the gravel was approximately VT. inch (12.7 mm), rather
than % inch (6.35 mm) in diameter (Figure 1). Again, at approximately 55 days, nitrite levels started to
increase only to decrease suddenly as a result of the accidental addition of chlorine to the system
(Figure 10). Nitrite  levels began to increase again at 95 days (Figure 10). Nitrite steadily increased to and
peaked at 0.2 mg N/L at 140 days. A similar decrease in ammonia through the contactor over the same
time period was observed. No nitrate  was produced in the contactor during this time.
                                                         Bed depth increase
                                       100       200       300      400
                                         Elapsed Time (days)

                        Figure 10. Nitrogen content of finished water from Contactor 2.
Bacterial acclimation and nitrification in Contactor 2 was slow. Considering variables that could impact
the process, the relatively high filter loading rate (flowrate) of 4.3 gpm/ft2(10.5 m/hr) (Figure 11)
through the contactor was thought to potentially be an issue. The loading rate was decreased to 3
gpm/ft2 (7.3 m/hr) then 1 gpm/ft2(2.4 m/hr) (Figure 11) between 140 and 160 days. The decrease in
loading rate resulted in an  immediate increase in nitrite production and a corresponding decrease in
ammonia to nearly 1 mg N/L by 200 days. No nitrate was produced in the contactor during this time.
Although some improvement was observed (i.e., ammonia oxidized), the progress was still very slow,
nitrite levels in the contactor were well over the 1 mg N/L MCL and reached nearly 2 mg N/L (Figure 10).
Between 200 and 280  days, there was a delayed generation of nitrate. The amount of nitrate formed
was very small and levels were generally less than 0.3 mg N/L.
                                                                                             19

-------
                       1000
                    e
                    S
                       400
                       200
                                                            Flow
                                                            Loading Rate
                                                                          En
                                                                          3"
                                                                       3   I
                                                                          •
                                                                          I
                          0       100      200       300       400      500
                                        Elapsed Time (days)
                               Figure 11. Contactor 2 flow and loading rate.


At approximately 290 days into the pilot study, the contactor loading rate was increased to 2.2 gpm/ft2
(5.4 m/hr) which did not appear to impact nitrogen  balances. Between 300 and 320 days, ammonia
levels in the contactor effluent nearly doubled to 2 mg N/L, nitrite levels decreased to 1 mg N/L and
nitrate fell back to near the detection limit. The changes in nitrogen distribution corresponded to a
steady decrease in oxygen levels in the contactor between 250 and 320 days. During this time, oxygen
decreased from approximately 10 mg O2/Lto 7 mg O2/L (Figure  12). Two backwash events at 320 days
and 350 days were necessary to clear the diffuser and bring oxygen levels back to approximately 10 mg
02/L
                                                                                            20

-------
                    ^  12


                    W
                    g  10
                    BO

                    O
                    •O  8
                     .
                    5
                    st
                    o
NH,
DC)
Betl depth increased
                               •	Diffuse blow-out
                               	  Diffuse! blow-out
0 •—"-
  0
                                            _L
              _L
                                  100
                       400
               200       300
           Elapsed Time (days)
Figure 12. Contactor 2 effluent ammonia and DO.
500
Operation of Contactor 2 paralleled that of Contactor 1, with the exception being that no
orthophosphate was added. Regardless of the changes made to Contactor 2 operation, satisfactory
results could not be achieved with respect to complete oxidation of ammonia to nitrate, which clearly
confirms the necessity of orthophosphate addition.
At Day 330, Contactor 2 was reconfigured such that its influent was Contactor 1 effluent rather than raw
water. At the time of reconfiguration, Contactor 1 was not able to achieve complete oxidation of all of
the source water ammonia all the way to nitrate. Reconfiguring the contactors essentially doubled
contactor bed depth. Also, shortly after the changeover, all the nitrogen leaving Contactor 2 was in the
nitrate form. Shortly after reconfiguration, however, Contactor 1 was able to completely oxidize all of
the ammonia to nitrate without the need for additional bed depth.
Ammonia, nitrite, and  nitrate levels entering Filter 2 were those exiting Contactor 2 (Figure 13 a).
Ammonia oxidation in the filter began at about 95 days (same as contactor) and nitrite increased
correspondingly to a very concerning 3.2  mg N/L by 190 days. As with the contactor, no nitrate was
formed up to this point. The filter loading rate during this time was between 1.2 gpm/ft2 (2.9 m/hr) and
1.5 gpm/ft2 (3.7 m/hr) up to 160 days (Figure 13b). It was reduced for a brief period to 0.5 gpm/ft2 (1.2
m/hr) which had no obvious impact on ammonia oxidation.  Nitrite levels decreased steadily to
approximately 1.8 mg  N/L by 260 days. During the same time, nitrate increased to 1.8 mg  N/L only to
decrease back to 0.2 mg N/L by 300 days were it remained for the rest of the pilot (Figure 13a). The
changes in nitrate appeared to correspond to changes in ammonia levels entering the filter and loading
rate changes.
                                                                                             21

-------
                                                                                           0.0
                                                                                  ion     wo
                      Elapsed Time (days)
                          Elapsed Time (days)
                             s
                             0*
                             51
                             >••
                             X
                             0
                             •3
                             a  6
      NH4
      DO
	  Beii depth mcieitse
	Diffib-ei blow-out
	  Dil'liiser blow-out
                                        100      200     300
                                             Elapsed Time (days)
                                                                       500
        Figure 13. Filter 2 data. A) Nitrogen content of finished water from filter 2, B) Filter 2 flowrate and hydraulic loading
        rate, and C) Filter 2 ammonia and oxygen.
Contactor 3/Filter3. Contactor 3 was also operated nearly identically to Contactor 1, with the exception
that orthophosphate was not added and the gravel was largest at 1 inch (25.4 mm). Given the problems
identified earlier, efforts to improve ammonia removal centered on Contactor I/Filter 1. Also, because
of the sample load to the laboratory, Contactor 3/Filter 3 was terminated at around 320 days. Trends in
nitrogen species through the contactor and filter were nearly identical to the Contactor 2/Filter 2
system, despite the larger media and will not be discussed here.
                                                                                                   22

-------
4.4 Removal of Iron from Source Water
Although the initial oxidation state of iron was not directly determined, it is reasonable to assume that
based on water chemistry, dissolved oxygen, and local geology that iron was initially in the reduced
Fe(ll) form. The oxygen level in the contactors and the pH of the source water led to rapid oxidation of
Fe(ll) to Fe(lll), and iron particles.
Iron in the source water, averaging 0.63 (±0.49 standard deviation) mg/L, was variable over the course
of the study as supported by the relatively high standard deviation (Table 3). It was not clear whether
the large range of iron levels was associated with the source variability, sampling issues, or analysis
issues associated with particulate-containing water samples. Although initially dissolved, it is also likely
that iron was in the particulate form at the time of or shortly after sampling. The presence of particles
could also cause variability.
Iron levels in the effluent of Contactors 1 and 2 were typically lower and generally within several tenths
of a mg Fe/L of the source water entering the contactors up until approximately 150 days of operation
(Figures 14 and 15). Interestingly, at approximately an event 150 days into operation rapidly triggered a
dramatic change in effluent iron levels in both contactors. After 150 days, iron in contactor effluents
became very sporadic, and some concentration spikes were very high. In some cases, iron levels were as
high as nearly 10 mg Fe/L. Orthophosphate addition was not associated with the cause because the
change in the iron pattern was observed to occur in both contactors at the same time. During about the
same time, loading rates were reduced and ammonia oxidation began to improve  in both systems. It
was during this time that the biological activity greatly increased and eventually thrived. It is believed
that iron particles became  incorporated into the biomass and were retained on the contactor. The
biofilm and iron sloughed off in an irregular manner which may have contributed to the occurrence of
sporadic and elevated iron spikes in the contactor effluents.
Regardless of the iron content in the contactor effluent, iron levels in filter effluent waters were very
low in dual media filters, essentially removing all of the iron (Figures 14 and 15). It was assumed that all
of the iron entering the filters was in the Fe(lll) or particulate form based on the oxygen concentration
and pH in the contactor water.
Iron removal through the filters was not impacted by filter loading rates (Figures 14 and 15). Filters were
operated between  0.5 gpm/ft2 (1.2 m/hr) and 2.2 gpm/ft2 (5.4 m/hr). Filter flowrates had to be lower
than contactor flowrate due to limitations in pilot design. At the completion of the study, Filter 1 was
operated at a loading rate of approximately 2 gpm/ft2 (4.9 m/hr).
                                                                                            23

-------
                        50     100    150    200    250    300    350
                                Elapsed Time (days)
Figure 14. Iron in raw water and treated water through Contactor 1 and Filter 1.
                        50     100    150    200    250    300    350
                                Elapsed Time (days)
Figure 15. Iron in raw water and treated water through Contactor 2 and Filter 2.
                                                                                             24

-------
4.5 Other Water Quality Parameters
Source water dissolved oxygen levels averaged 3.6 ± 1.1 mg/L over the course of the study (Figure 16).
The source water temperature averaged 14.8 ±2.2° C over the course of the pilot and did not change
through the pilot system. Given the source water was a ground water; it remained relatively stable
through the study period. Water temperature was not impacted by outside air temperature, which, at
times, could be very cold (Figure 16). The pilot study demonstrated that biological treatment will work in
colder regions, provided groundwater is the source of drinking water and the facility is adequately
heated. TOC in the source water averaged 1.3 mg C/L and did not change through the pilot system
contactors and filters.
                                                *_  -
                                                8  o
                                                f

                        30
                        25
                     O

                     f
                     a<
                     •a
                     to
                     j;  10
                     i
                     y;
                     S
DO
Teinp
An Temp
                                                                     25
                                                                     20
               15
                   a
                   a.
               10   £
                   
-------
             360
         a
         c-
             350
             345
        2   340
        o
             335
             330
                           100         200         300

                                    Elapsed Time (days)
                                                               400
         Figure 17. Total alkalinity of raw, Contactor 1 effluent and Filter 1 effluent.
                                100      200      300       400       500

                                     Elapsed Time (days)


Figure 18. Heterotrophic plate counts (HPCs) in raw, Contactor 1 effluent and Filter 1 effluent.
                                                                                                    26

-------
                          u
                           a

                           £ i
-------
5.   Discussion and Summary
5.1 Discussion
The pilot study demonstrated the ability of biological treatment to effectively remove ammonia and iron
from the source water. The development of biological activity and subsequent complete oxidation of
ammonia to nitrate in the system was initially much slower than anticipated based on previous work,
although the site's water quality was more challenging from an ammonia level standpoint than
conditions in past work. Fortunately, the pilot study was valuable in identifying key reasons for the
discrepancies, and more importantly, identifying engineering and design improvements to address
them. For example, loading rate targets, the sensitivity of the system to dissolved oxygen throughout
the contactor, need to keep the diffuser clean, occasional backwash of contactor, and phosphate feed
were all identified as  important.
              Table 5. Final Design and Operating Parameters
Parameter
Filter loading rate
m/hr
gpm/ft2
Airflow/rate
L/min
cfm/ft2
Backwash conditions
duration, min
bed expansion, %
m/hr
gpm/ft2
Contactor
depth, cm
depth, inches
effective size, mm
effective size, inches
Filter
anthracite depth, cm
anthracite depth, inches
anthracite mm
anthracite , inches
sand depth, cm
sand depth, inches
sand, mm
sand, inches
Contactor

5.4(1.2-10.5)
2.2(0.5-4.3)

2.5
2.86

5
0
124
51

76.2
30
12.7(6.35-31.8)
0.5(0.25-1.25)



--
—


--
--
Filter

4.9(1.22-5.4)
2.0(0.5-2.2)

—
—

15
50
41.5
17

-
--
—
—

50.8
20
0.97
0.04
25.4
10
0.45
0.018
                                                                                        28

-------
By the termination of the pilot study, complete oxidation of the source water ammonia (3.2 mg N/L) to
nitrate was achieved in Contactor 1 and removal of iron (0.63 mg Fe/L) through the anthracite/sand
filter followed. Other operating and maintenance parameters are summarized  in Table 5.
Orthophosphate addition was necessary and it is uncertain how long it would have taken to get the
system to the same goal, had all of the parameters been optimized at the start-up.

5.2 Summary of Key Findings
The biological treatment ammonia pilot study produced a number of new and very important findings
that will improve the drinking water field's understanding of biological water treatment in general and
how to effectively operate such systems. The following findings are highlighted:
•   Once optimized, the biological pilot system achieved the treatment goal of completely oxidizing all
    of the ammonia in the source groundwater to nitrate. Complete oxidation of ammonia all the way to
    nitrate was eventually achieved in the contactor (Contactor 1) that contained 30 inches (76.2 cm) of
    small gravel. The addition of air at the base of the contactor was necessary design feature to address
    the oxygen demand of the nitrification process and iron oxidation.
•   A dual media (20 inches [50.8 cm] anthracite/10 inches [25.4 cm] sand) filter (Filter 1) after the
    contactor provided additional ammonia/nitrite oxidation, and achieved excellent and consistent iron
    removal.
•   The source water contained very little phosphorus. Orthophosphate is an important biological
    nutrient and its addition was necessary to increase the rate of microbial acclimation, particularly
    with regards to nitrite oxidizing bacteria. The system responded almost immediately to the addition
    of Orthophosphate. A dose of 0.3 mg PO4/L was used in the pilot. The Orthophosphate feed was
    terminated for an extended period of time during which  no negative impact on the system's
    performance was noted.
•   Maintaining near saturated dissolved oxygen levels in the contactor was critical to the processes'
    operation and effectiveness at achieving desired ammonia  oxidation and iron removal. Drop in
    dissolved oxygen levels due to diffuser "clogging"  resulted  in delayed oxidation of ammonia in the
    contactor and release of nitrite. Dissolved oxygen  monitoring was a good process measurement tool
    and must be incorporated into full-scale operation. Diffuser design will also be very important
    engineering aspect of the full-scale system.
•   Contactor and filter loading rates were important  operating variables, although the pilot system was
    more sensitive to Orthophosphate and oxygen concentration. The pilot demonstrated that a
    contactor and filter operated in series at loading rates of 2.2 gpm/ft2 (5.4 m/hr) and 2.0 gpm/ft2 (4.9
    m/hr), respectively, met desired finished water quality objectives.
•   Alkalinity decrease following nitrification in the systems was predicted by theoretical considerations
    and could be used as an additional process monitoring tool.
•   Contactor maintenance was minimal. Although, not systematically evaluated during the pilot, there
    was some evidence to suggest backwashing an acclimated  contactor was beneficial. As a result,
    monthly backwash of the contactors is recommended. Similarly, minimal filter maintenance was

                                                                                           29

-------
    necessary. Filters were backwashed only once a week by achieving 50% bed expansion for 15
    minutes.

5.3 Future Work/Questions
Fortunately the pilot was extremely successful in that it identified many new and important details
regarding the operation of a biological ammonia oxidation system that would not have been identified
otherwise. The time it took to make the discoveries and modifications to address them, however,
extended the length of the pilot well beyond what was initially expected. In addition, there was not
enough time to perform some of the planned investigations. As a result, a number of questions
regarding system operation and optimization still remain. Specifically;

•   How long would the pilot system (and eventually full-scale system) take to acclimate had it been
    operated from the start-up under the "optimum" conditions operated at the termination of the pilot
    study? How would the corresponding nitrate and nitrate contactor and filter finished water profiles
    look? How long and at what concentration would nitrite peak at? How would the hours of operation
    (hours per day) impact acclimation period?

•   There was not a scientific basis behind the orthophosphate dose selected in the pilot and the dose
    used was well above the stoichiometric amount necessary for bacterial cell growth in such a system.
    Relevant questions to orthophosphate dosing include: What is the optimal orthophosphate dose? Is
    there a benefit to a start-up dose to get the system going? If so, what is the minimum maintenance
    dose to keep the system going? The system was not impacted by short-term (4 weeks)
    orthophosphate feed breaks but how will the system perform under long-term orthophosphate
    stoppages?

•   The addition  of chlorine to the system after the contactor would eliminate the ability of the filter to
    oxidize ammonia and nitrite, and the safety factor. In such a case, contactor bed depth could be
    increased to provide an additional safety margin. Related questions would be "What are nitrogen
    profiles through a contactor as a function of bed depth?" and "Can chlorinated water be used to
    backwash the filters/contactors and yet still maintain microbiology of the systems, and nitrification
    capability?"

•   What is the relationship  between bed depth, media, loading rate, and ammonia oxidation? The pilot
    design and operation was based on  past EPA work and are within "typical" ranges of granular media
    drinking water treatment systems.
                                                                                           30

-------
6.   References
Andersson, A.; Laurent, P.; Kihn, A.; Prevost, M.; Servais, P. Impact of Temperature on Nitrification in
Biological Activated Carbon (BAG) Filters used for Drinking Water Treatment. Water Research. 2001,
35:2923-2934

Bremer, P.J.; Webster, B.J.; Wells, D.B. Biocorrosion of Copper in Potable Water. Journal AWWA. 2001,
93, 82-91

de Vet, W.W.J. M; van Loosdrecht, M.C.M.; Rietveld, L.C. Phosphorus Limitation in Nitrifying
Groundwater Filters.  Water Research. 2012, 46:1061-1069

Fleming,  K.K.; Harrington, G.W.; Noguera, D.R. Nitrification Potential Curves: A New Strategy for
Nitrification Prevention. Journal A WWA. 2005, 97, 90-99

Lee, S.H.; O'Connor, J.T.; Banerji, S.K. Biologically Mediated Corrosion and Its Effects on Water Quality in
Distribution Systems. Journal AWWA. 1980, 72, 636-645

Lytle, D.A.;  Sorg, T.J.,  Wang, L.;  Muhlen, C.; Rahrig, M.; French, K. Biological Nitrification in Full-Scale and
Pilot-Scale Iron Removal Drinking Water Treatment Plant Filters. Jour.  Water Supply: Res. and Tech.:
AQUA. 2007, 56, 125-136

Odell, L.H.;  Kammeyer, G.J.; Wilczak,  A.; Jacangelo, J.G.; Marcinko, J.P.; Wolfe, R.L Journal AWWA
Occurrence of Nitrification in Chloraminated Distribution Systems. 1996, 88, 74

Rittman,  B.E.; Snoeyink, V.L Achieving Biologically Stable Drinking Water. Journal A WWA. 1984.76, 106

Scherrenberg, S. M.; Menkveld, H.W.; Bechger, M.; van der Graaf, J.H. Minimising the risk on
phosphorus limitation; an optimised  coagulant dosage system. Water Sci Technol. 2011, 64:1708-1715.

Scherrenberg, S. M.; Neef, R.; Menkveld, H.W.; van Nieuwenhuijzen, A.F.; van der Graaf, J.H.
Investigating Phosphorus Limitation in a Fixed Bed Filter with Phosphorus and Nitrogen Profile
Measurements. Water Environ Res. 2012, 84:25-33.

Shammas, N.K. Interactions of Temperature, Ph, and Biomass on the Nitrification Process. J. Water
Pollut. Control Fed. 1986. 58, 52-58.

Suffet, I.H.; Corado, A.; Chou, D.; McGuire, M.J.; and Butterworth, S. AWWA Taste and Odor Survey.
Journal AWWA. 1996. 88, 168-180.

U.S. Environmental Protection Agency (US EPA), Process Design  Manual for Nitrogen  Control,
Technology Transfer,  U.S. EPA, Washington D.C., 1975.

U.S. Environmental Protection Agency (US EPA), Manual: Nitrogen Control. Office of Research and
Development. Office  of Water, Washington, DC  EPA1625/R-93/010 1993.

WHO, Ammonia in Drinking-Water. Background  Document for Preparation of WHO Guidelines for
Drinking-Water Quality. Geneva, World Health Organization  (WHO/SDE/WSH/03.04/1) 2003.
                                                                                            31

-------
APPENDIX A
                     30
                 9
                 o
                 •3
                 O)
                     10
                              J_
                                                • DO
                                                • Temp
                                                     J_
                              100     200      300      400

                                  Elapsed Time (days)

                           A 1. Contactor 1 effluent temperature and DO.
                                                              30
                                                              25
                                                              20
      =
  15   a
      ~
      e.
      ~
  10   H
500
                                                                                  32

-------
    30
                                                         30
cu
(Si
>-.


O

5
    25
15
    10
                                       • DO

                                       - Tenp
                                                         25
                                                         20
                                                     15
                                                         10
O


~
c:

5



3
H
                J_
                                              J_
                100       200       300       400


                      Elapsed Time (days)


            A 2. Contactor 2 effluent temperature and DO.
                                                      J—I 0

                                                       500
    30
                                                         30
0
hj
efc
c
O)
6
    15
    10
                                                         25
                                                     20
                                                         15
                                                         10
                                                              u
                                                              O
                                                              o
                                                              u
                                                          I
                                                          H
                J_
                                              _L
                100       200       300       400


                      Elapsed Time (days)


          A 3. Contactor 2 and Filter 2 effluent Temp and DO.
                                                       500
                                                                                  33

-------
             100        200        300       400
                    Elapsed Time (days)
              A 4. Nitrogen content of raw water.
500
                    100       200       300
                       Elapsed Time (days)
                                                400
       A 5. Nitrogen content of finished water from Filter 3.
(Image was scanned from a hard copy. Resolution will be improved.
                                                                               34

-------