Nitrogen Transformation and Recovery

Source: Environmental Biotechnology: Principles and Applications, 2nd Edition
ISBN: 9781260441604
Authors: Bruce E. Rittmann Ph.D., Perry L. McCarty Sc.D.
13. Nitrogen Transformation and Recovery

The plant nutrients nitrogen (N) and phosphorus (P) are essential for the production of food to sustain the world’s population,
but they become undesirable pollutants when discharged to rivers, streams, lakes, and groundwater. In surface waters, they
stimulate the growth of unwanted algae and other aquatic plants, a phenomenon called eutrophication. Ammonium discharge
also can result in serious oxygen depletion when it is nitrified to nitrate, and nitrate can be harmful to infants and animals
drinking the water. Thus, plant nutrients represent another example of materials that may be beneficial or harmful, depending
upon location and use. This chapter is concerned with nitrogen and biological processes that can be used to transform harmful
nitrogen forms in wastewaters into nondetrimental materials, such as dinitrogen (N 2 ) gas. Biological processes for control of
phosphorus are discussed in Chapter 14.
Traditionally in wastewater treatment, the bacterial process of ammonium oxidation (or nitrification) is used to transform
ammonium nitrogen to the more oxidized forms, nitrite and nitrate. If needed or desired, the oxidized forms can then be reduced
into dinitrogen gas (N2 ) through biological denitrification, sending the N back to the atmosphere from which it came. Starting
just before the advent of the twenty-first century, new transformation processes were discovered for nitrogen transformation,
and their use may permit more sustainable approaches for nitrogen removal from wastewater. Newly included are the
anaerobic ammonia-oxidizing or anammox organisms (Winkler et al., 2012), the nitrifying organisms of the Archaea domain
(Stahl and de la Torre 2012), the nitrifying bacteria that carry out ammonium oxidation all the way to nitrate, the denitrifying
microorganisms that can use methane as an electron donor (Lee et al., 2018), and the microorganisms that oxidize nitrite to
produce nitrous oxide N 2 O (Weissbach et al., 2018), a powerful greenhouse gas that might also be used as a high-energy fuel.
It is necessary to know and understand not only the traditional methods, but also the newer organisms and the processes they
govern, as well as research that is underway toward making all approaches more sustainable. This chapter puts more emphasis
on the traditional approaches, but some of the newer methods also are included, particularly the anammox process.
This chapter begins with a review of the several forms of nitrogen and their significance as pollutants. Then, the various
nitrogen transformation reactions are developed. This material lays the foundation for the subsequent sections on the different
nitrogen-transformation processes.
The second part of this chapter provides a more detailed description of the microbiological processes involved in nitrogen
transformations, together with the key biochemical and physiological characteristics of the prokaryotic organisms involved.
First to be addressed is nitrification of ammonium to nitrite and nitrate. Second is denitrification of oxidized nitrogen species
into harmless dinitrogen gas. Of special note is the discussion of the relatively new anammox denitrification process. The
different nitrogen-transformation processes can be combined in various ways to achieve nitrogen removal from wastewaters,
mostly as dinitrogen gas. In each of these nitrogen-transformation processes, some ammonium also is converted into cellular
organic nitrogen, which can be removed from the main stream by settling or filtration.
The final section of this chapter contains a mass-balance comparison of different nitrogen-treatment schemes that currently
are or potentially could be used under different circumstances. This comparison highlights the relative advantages and
disadvantages of each, including their resource requirements.
13.1. Nitrogen Forms, Effects, and Transformations
© McGraw-Hill Education. All rights reserved. Any use is subject to the Terms of Use, Privacy Notice and copyright information.
Figure 13.1 illustrates the nitrogen cycles of importance in nature and environmental biotechnology. Kuypers et al. (2018) have
provided a detailed summary of the current understanding of nitrogen cycles. Nitrogen has eight oxidation states, varying from
−3 for ammonium (NH+ 4
) to +5 for nitrate (NO− −
3 ). Other common oxidation states are +3 for nitrite (NO2 ), which is formed
either through nitrification or denitrification, and −3 for organic nitrogen (orgN) produced during cell synthesis. Another species
formed either through nitrification via +1 nitroxyl (HNO) or denitrification via +2 nitric oxide (NO) is the +1 nitrous oxide (N2 O), a
powerful greenhouse gas with a heat warming potential about 265 times that of carbon dioxide (IPCC, 2013). About 78% of the
Earth’s atmosphere is the zero-oxidation-state dinitrogen (N2 ) gas, an inert form that is not directly useful for the growth of
plants. However, many different microbial species also can convert N2 into ammonium or nitrate, the most useful forms for
plant growth, through the process of nitrogen fixation. Included here are nitrogen-fixing bacteria, such as the photosynthetic
cyanobacteria and the Azotobacter organisms, which are associated with some plants, especially legumes. Some atmospheric
nitrogen also can be fixed through high-temperature combustion or by lightening into the +2 nitric oxide (NO) and the +4
nitrogen dioxide (NO2 ). Finally, hydroxylamine (NH2 OH) is an important intermediate in the nitrification process.
Figure 13.1 Nitrogen transformation pathways and oxidation states. (Source: Kuypers et al. (2018).)
Because of today’s large world population, natural nitrogen-transformation processes are insufficient to adequately supply the
world with the nitrogen forms it demands for food production. A chemical synthesis method, the Haber–Bosch process, which
transforms N2 into ammonium, was invented to help supply the need. However, the Haber–Bosch process is energy intensive,
requiring about 7% of the natural gas used in the world today (McCarty et al. 2011). For this reason, it is highly desirable to make
direct use of wastewaters’ ammonium and organic nitrogen content for agriculture production, thus reducing the need for
Haber–Bosch-produced ammonium nitrogen for fertilizer, as well as the energy required to carry out the nitrification and
denitrification processes to convert ammonium back to N2 .
When discharged to surface waters, the usual fate of treated wastewaters, the reduced or oxidized forms of nitrogen stimulate
the growth of unwanted aquatic plants and other undesirable effects associated with eutrophication. In such cases, and if
nitrogen capture is not economical, conversion of reduced and oxidized forms of nitrogen to N2 gas may be justified. But the
planners and designers of wastewater-treatment facilities need to be well informed of all environmental impacts of nitrogen
discharges and treatment. No single solution is appropriate for all cases, and traditional approaches may not be the most
sustainable from the perspective of environmental and economic impacts.
Nitrogen is an element that poses many important concerns as well as benefits for agriculture. The most prevalent nitrogen
forms that the water engineer or scientist confronts are ammonia, nitrite, nitrate, and organic N. Of growing importance is the
unwanted production of nitrous oxide, because of its significant global-warming potential. Other environmental concerns are
NO and NO2 , collectively called NOx, which contribute to smog and are formed during combustion of fuel in automobiles, power
plants, and industrial complexes.
13.2. Nitrogen’s Transformation Reactions
Many nitrogen transformations are of particular interest in water-quality engineering. Here, we provide a summary of the major
stoichiometric reactions associated with these transformations, reactions that will be used in the following sections on
nitrification and denitrification. Included first in Table 13.1 are the important half-reactions. These half-reactions are then
combined in the usual manner, as outlined in Table 13.2, to form each of the complete biological reactions of interest for
nitrogen transformations. The energy portion of each equation of interest [fe(Rai − Rdi)] is first written and added to the
synthesis portion [fs(Rci − Rdj)] using the appropriate equations from Table 13.2. Most often, the Rdi for the energy and
synthesis half-reactions is the same, but not for the anammox overall reaction. With the autotrophic anammox organisms,
nitrite, not ammonium, is the reductant for converting CO 2 to cell carbon. Thus, for the synthesis portion of the anammox
reaction, Rd4 must be used (Lee et al., 2013).
Table 13.1 Half-Reactions Associated with Biological Transformations of Nitrogen Species
Equation Description Reduction Equation
Source: Adapted from the supporting information in McCarty (2018).
Electron Donor Half-Reactions
Rd 1 Ammonium oxidation to nitrate

1 5 1 3
NO–3 + H+ + e– = NH +
4
+ H2 O
8 4 8 8
Rd 2 Ammonium oxidation to nitrite

1 4 1 1
NO–2 + H+ + e– = NH +
4
+ H2 O
6 3 6 3
Rd 3 Ammonium oxidation to N 2
1 4 1
N2 + H+ + e– = NH +
4
6 3 3
Rd 4 Nitrite oxidation to nitrate

1 1 1
NO–3 + H+ + e– = NO−
2 + H2 O
2 2 2
Rd 5 Organic oxidation of methanol

1 1 1
CO2 + H+ + e– = CH3 OH + H2 O
6 6 6
Rd 6 Hydrogen (H 2) oxidation
1
H+ + e– = H2
2
Electron Acceptor Half-Reactions
Ra1 Aerobic reactions

1 1
O2 + H+ + e– = H2 O
4 2
Equation Description Reduction Equation
Ra2 Nitrate reduction to N 2

1 6 1 3
NO–3 + H+ + e– = N2 + H2 O
5 5 10 5
Ra3 Nitrite reduction to N 2

1 4 1 2
NO–2 + H+ + e– = N2 + H2 O
3 3 6 3
Ra4 Methanogenesis
1 1
CO2 + H+ + e– = CH4 + H2 O
8 8
Synthesis Half-Reactions
Rc1 Ammonium as N source

1 1 1 1 9
CO2 + HCO–3 + NH +
4
+ H+ + e– = C5 H7 O2 N + H2 O
5 20 20 20 20
Rc2 Nitrate as N source

4 1 1 1 1 13
CO2 + HCO–3 + NO− + –
3 +H +e = C5 H7 O2 N + OH− + H2 O
28 28 28 28 28 28
Rc3 Nitrite as N source

4 1 1 + − 1 2 10
CO2 + HCO–3 + NO−
2 +H +e = C5 H7 O2 N + OH− + H2 O
26 26 26 26 26 26
Table 13.2 Half-Reactions Used for Constructing Energy and Synthesis Reactions
Energy Half-Reactions Synthesis Half-Reactions
Reaction Ra i Rdi Rc Rdj
Source: Adapted from the supporting information in McCarty (2018).
Aerobic organic oxidation Ra1 Rd 5 Rc1 Rd 5
Organic methanogenesis Ra4 Rd 5 Rc1 Rd 5
NH +
4
nitrification to NO−
2 Ra1 Rd 2 Rc1 Rd 2
−
NO−
2 nitrification to NO3 Ra1 Rd 4 Rc1 Rd 4
Heterotrophic denitrification with NO−

3 Ra2 Rd 5 Rc1 Rd 5
Heterotrophic denitrification with NO−

2 Ra3 Rd 5 Rc1 Rd 5
Autotrophic denitrification with NO−

3 Ra2 Rd 6 Rc2 Rd 6
Autotrophic denitrification with NO−

2 Ra3 Rd 6 Rc2 Rd 6
Anammox Ra3 Rd 3 Rc1 Rd 4
One example to illustrate the linking of the energy and synthesis half-reactions is that for aerobic oxidation of one electron
equivalent (8 g COD) of organic matter in the form of methanol (CH3 OH):
1 1 1 1 1
fe(Ra1 − Rd5) = fe [ O2 + H+ + e− = H2O] + fe [ CH 3OH + H2O = CO2 + H+ + e− ]
4 2 6 6 6
1 1 1 1
= fe [ CH 3OH + O2 = CO2 + H2O]
6 4 6 3
(13.1)
1 1 1 1 9
fs(Rc − Rd1) = fs [ CO2 + HCO− 3 + NH+ + H +
+ e− = C 5H 7O2N + H O]
5 20 20 4 20 20 2
1 1 1
+fs [ CH 3OH + H2O = CO2 + H+ + e− ]
6 6 6
⎡ 1 CH 3OH + 1 CO2 + 1 HCO− + 1 NH+ ⎤
= fs ⎢
⎢ ⎥
⎥
3 4
6 30 20 20
⎣= ⎦
1 9
C 5H7O2N + H2O
20 20
(13.2)
Adding Equations 13.1 and 13.2 together generates the reaction for complete aerobic organic oxidation:
1 f f f f
CH 3OH + e O2 + s CO2 + s HCO–3 + s NH+
4
6 4 30 20 20
9f
= s C 5H7O2 + e CO2 + ( e + s ) H2O
f f f
20 6 3 20
(13.3)
Here, methanol is just one example of an organic donor. One-sixth mole of methanol has a COD of 8 g, as does one electron
equivalent of any electron donor. Thus, for any single or mixture of organic compounds, 8 g COD or BODL rather than 1/6 mol
can be used.
Although the values for CO2 , HCO− 3 , and H2 O may be important for alkalinity and pH effects, they are not of interest in terms
of electron and N equivalents. Dropping CO2 , HCO− 3 , and H2 O from Equation 13.1 leads to a skeleton relationship (not fully
balanced) that contains only the stoichiometry of the variables of interest in electron- and N-equivalent calculations for aerobic
organic oxidation:
Aerobic organic oxidation skeleton equation:
1 f f fs
g COD + e O2 + s NH+
4
= C 5H7O2N
8 4 20 20
(13.4)
Here, O2 has a molecular weight of 32 g, NH+

4
-N 14 g, and C5 H7 O2 N for cell VSS 113 g.
Skeleton equations derived for other microbiological reactions are (for 1 e eq of donor):
1 f f f
Organic methanogenesis: g COD + s NH+
4
= e CH 4 + s C 5H7O2N
8 20 8 20
(13.5)
1 1
Ammonium oxidation to nitrite: ( + s ) NH+
f f f
4
+ e O2 = NO–2 + s C 5H7O2N
6 20 4 6 20
(13.6)
NO–2 + e O2 + ( s ) NH+
fe f f f f
Nitrite oxidation to nitrate: 4
= e NO–3 + s C 5H7O2N
2 4 20 2 20
(13.7)
1 f f
Heterotrophic denitrification with nitrate: g COD + e NO–3 + s NH+
4
8 5 20
f f
= e N2 + s C 5H7O2N
10 20
(13.8)
1 f f
Heterotrophic denitrification with nitrite: gCOD + e NO–2 + s NH+4
8 3 20
f f
= e N2 + s C 5H7O2N
6 20
(13.9)
1
gH2 + ( e + s ) NO−
f f
Autotrophic denitrification with nitrate: 3
2 5 28
1 f
= ( e + s ) N2 + s C 5H7O2N
f f
2 5 28 28
(13.10)
1
gH2 + ( e + s ) NO−
f f
Autotrophic denitrification with nitrite: 2
2 5 26
1 f
= ( e + s ) N2 + s C 5H7O2N
f f
2 5 26 26
(13.11)
( + ) NH+ + ( e + s ) NO–2
fe fs f f
Anammox reaction: 4
3 20 3 2
f f f
= e N2 + s NO–3 + s C 5H7O2N
3 2 20
(13.12)
Equations 13.4 through 13.12 can be solved after selecting appropriate values forfs0, b, and θx for any nitrogen-transformation
case under consideration. Values for fe and fs can be determined as outlined in Chapter 5:
1 + 0.2bθ x
fe = 1 − fs, fs = fs0
1 + bθ x
(13.13)
The other values can be estimated as outlined in Chapter 6.
Example
Example 13.1: Analysis of a Reactor for Heterotrophic Denitrification
A wastewater with a nitrate-N concentration of 45 mg/l is fed to a denitrification bioreactor. What inlet concentration of
BOD L is needed for complete reduction of nitrate to N2 ? Assuming ammonium is being used for biomass synthesis, what
concentration is needed, and what will be the concentration of biomass resulting from the transformation? Assume the
value of fs0 is 0.50, b is 0.05 d−1, and θx is 8 d.
Equation 13.8 is the appropriate one to use for this example. Needed first is to determine the values to use forfe and fs,
which can be found using Equation 13.13:
1 + 0.2(0.05)(8)
fs = 0.5 = 0.39 and fe = 1 − 0.39 = 0.61
1 + 0.05(8)
Substituting these values into Equation 13.8 yields the skeleton reaction:
1 0.61 0.39 0.61 0.39

BODL + NO−
3 + NH+
4
= N2 + C 5H7O2N
8 5 20 10 20
Thus, the concentration of BODL required for denitrification is
64 mg BODL 45 mg NO–3-N mmol NO–3-N 5 e meq NO–3-N

BODL = ( )( ) ( ) = 210 mg/l
8 e meq l 14 mg NO–3-N 0.61 mmol
The concentration of ammonium-N required for biomass synthesis is
0.39(14) mg NH+ -N 45 mg NO–3-N mmol NO–3-N 5 e meq NO–3-N

NH+ -N = 4
( )( ) ( )
4 20 e meq l 14 mg NO–3-N 0.61 mmol
= 7.2 mg/l
The concentration of VSS formed is:
)( )(
0.39(113) mg VSS 45 mg NO–3-N mmol NO–3-N 5 e meq NO–3-N
VSS = [ ]( )
20 e meq l 14 mg NO−
3 -N
0.61 mmol
= 58 mg/l
Example
Example 13.2: Analysis of a Reactor for Autotrophic Denitrification
Repeat Example 13.1 using autotrophic denitrification with H2 gas as the electron donor. The input nitrate-N concentration
remains 45 mg/l, and we assume that NO− 3 is the electron acceptor and nitrogen source. What inlet concentration of H2 is
needed for complete reduction of nitrate to N 2 ? What will be the concentration of biomass resulting from the
transformation? Assume the value of fs0 is 0.20, b is 0.02 d−1, and θx is 8 d.
Equation 13.10 is the appropriate one to use for this example. Needed first is to determine the values to use forfe and fs,
which can be found using Equation 13.13:
1 + 0.2 (0.02) (8)

fs = 0.2 = 0.18 and fe = 1 − 0.18 = 0.82
1 + 0.02 (8)
Substituting these values in Equation 13.10 yields the skeleton reaction:
1
H + 0.170 NO−
3 = 0.082 N2 + 0.00643 C 5H7O2N
2 2
Thus, the concentration of H2 required in the influent flow for denitrification is
1 mg H2 45 mg NO–3-N mmol NO–3-N 1 e meq

H2 = ( )( – )( ) = 19 mg/l
e − meq l 14 mg NO3-N 0.170 mmol
The concentration of VSS formed is
45 mg NO−
]( )(
0.00643(113) mg VSS 3 -N mmol NO–3-N 1 e meq
VSS = [ – )( ) = 14 mg/l
e − meq l 14 mg NO3-N 0.170 mmol
We see that the net biomass production is about one-fourth of that for heterotrophic denitrification; this is a result of the
lower fs0 for autotrophy and the need to reduce NO−3 as the N source. The corollary is that the BODL equivalent of the
delivered H2 also is less—(19 mg H2 /l) (1 e meq/mg H2 ) (8 g BODL/e meq) = 152 mg BODL/l—because less donor has to be
supplied to support less biomass synthesis.
Example
Example 13.3: Nitrification in an Aerobic Bioreactor
A 100-m3 aerobic activated sludge reactor is operated to remove soluble BODL and to oxidize ammonium to nitrate. The
influent wastewater has a BODL of 250 mg/l and contains 40 mg/l of ammonium-N. If the reactor is operated with
θ = 0.25 d and θ x = 12 d, what mass of O2 is consumed and how much total biomass (VSS) is produced each day?
Assume removal is 98% for BOD and 95% for ammonium. Also assume that fs0 and b are 0.6 and 0.10 d−1, respectively, for
BOD removal and 0.12 and 0.05 d−1 for ammonium oxidation.
This question can be answered using Equations 13.4 and 13.6. But let’s start by determining fs and fe for each of the
reactions.
1 + 0.2(0.10)(12)
BODL: fs = 0.6 = 0.34, fe = 1 − fs = 0.66
1 + 0.10(12)
1 + 0.2(0.05)(12)
Ammonium: fs = 0.12 = 0.084, fe = 1 − fs = 0.916
1 + 0.05(12)
Next, we evaluate BODL removal using Equation 13.4, since some of the ammonium will be used for heterotrophic cell
synthesis and will not be available for nitrification. We need to learn how much that is by writing Equation 13.4 for 8 g COD
(or 1 e eq):
0.66(32 g) 0.34(14 g) 0.34(113 g)

8 g COD + O2 + NH+
4
= C 5H7O2N
4 20 20
Then, considering BODL removal efficiency is 98%:
250 mg BODL 1 0.66(32 g O2)

O2 consumption = 0.98 ( )( )( ) = 162 mg/l
l 8 g BODL 4
250 mg BODL 1 0.34(113 g VSS)

Biomass (VSS) synthesis = 0.98 ( )( )( ) = 59 mg/l
l 8 g BODL 20
0.34(14 g NH+ -N)

)( ) = 7.3 mg/l
250 mg BODL 1
NH+
4
-N usage = 0.98 ( )( 4
l 8 g BODL 20
The ammonium-N available for nitrification equals the reactor input minus that used for heterotrophic cell synthesis, (40.0 −
7.3) = 32.7 mg/l.
Substituting appropriate values into Equation 13.6, we obtain
1 0.084 0.916 1 0.084

( + ) NH+
4
+ O2 = NO−
3 + C 5H7O2N
8 20 4 8 20
or 0.129 NH+
4
+ 0.229 O2 = 0.125 NO−
3 + 0.0042 C 5H7O2N
Then, considering that ammonium removal efficiency is 95%:
32.7 mg NH+ -N
O2 consumption = 0.95 ( )[ ] = 126 mg/l
4 0.229(32) g O2
l 0.129(14) g NH+
4
-N
32.7 mg NH+ -N
VSS synthesis = 0.95 ( )[ ] = 8.2 mg/l
4 0.0042(113) g VSS
l 0.129(14) g NH+
4
-N
Finally, the answers to the questions asked are
)( 3 ) = 115 kg/d
(162 + 126) mg O2 100 m3 kg ⋅ l
Total O2 consumed = [ ](
l 0.25 d 10 mg ⋅ m3
)( 3 ) = 26.8 kg VSS/d
(58.8 + 8.2) mg VSS 100 m3 kg ⋅ l
Total biomass formed = [ ](
l 0.25 d 10 mg ⋅ m3
13.3. Nitrification
−
Nitrification is the microbiological oxidation of NH+
4
-N to NO−
2 -N and NO3 -N. Because of the oxygen demand (up to 4.57 g
O2 /g NH+
4
-N) and toxicity to aquatic microorganisms, NH+ 4
-N removal is a mandated process for many wastewaters. In
−
addition, wastewater treatment that involves denitrification of NO− 2 -N or NO3 -N first requires the nitrification of NH4 -N to
+
those nitrogen forms. Nitrification for removing NH+ 4

-N from drinking-water supplies also may be practiced to make the water
biologically stable or to eliminate the free-chlorine demand that produces chloramine when free chlorine is desired (Rittmann
and Snoeyink 1984).
Reviewed first are the key biochemical and physiological characteristics of the unique bacteria that perform nitrification. These
characteristics are responsible for the process design features that are essential for successful nitrification. Then, suspended
growth and biofilm nitrification are described.
13.3.1. Biochemistry, Physiology, and Kinetics of Nitrifying Bacteria

The nitrifying bacteria are autotrophs, chemolithotrophs, and obligate aerobes. Knowledge of each factor is crucial for
understanding when nitrifiers can be selected and accumulated in a biological process. Being autotrophs, the nitrifiers must fix
and reduce inorganic carbon. This is an energy-expensive process that is primarily responsible for nitrifiers having much
smaller values of fs0 and Y than do the aerobic heterotrophs that always populate biological treatment systems. Their
chemolithotrophic nature makes fs0 and Y still smaller, because their nitrogen electron donors release less energy per electron
equivalent than do organic electron donors or H2 . Of course, the low Y value translates into a small maximum specific growth
rate (μ̂ ) and a large θ min
x . Thus, nitrifiers are slow growers.
Since a report by Konneke et al. (2005), microorganisms oxidizing ammonium are now known to include representatives of the
Archaea domain [ammonium-oxidizing archaea (AOA)], along with many members of the Bacterial domain [ammonium-
oxidizing bacteria (AOB)]. In both cases, they are obligate aerobes and use O2 for respiration.
Nitrification has generally been considered a two-step process carried out by distinct nitrifying species. In the first step,NH+
4
is
oxidized to NO− 2 (nitritification) by microorganisms from either of the domains. As indicated in Figure 13.1, the first step in this
overall oxidation is the conversion of ammonium to hydroxylamine (NH2 OH) by the enzyme ammonium monooxygenase
(AMO). While oxidation to NH2 OH is an energy-generating reaction, it is one that is biologically difficult to accomplish, and so
microorganisms do this with AMO. Because monooxygenation is an electron- and energy-requiring activation reaction, energy
must be added to the reaction rather than being obtained from it. That energy is obtained from NADH through its oxidation to
NAD+:
NH+
4
+ O2 + NADH = NH2OH + NAD+ + H2O
(13.14)
This means that no electron equivalents are generated by the oxidation of ammonium to hydroxylamine; in fact, two electron
equivalents are consumed.
The next steps, oxidation to nitric oxide and then to nitrite, are the important energy-generating reactions that allow the
ammonium oxidizers to grow. The first step is oxidation to nitric oxide, which captures the energy released to form NADH:
NH2OH + 1.5 NAD+ = NO + 1.5 NADH + 1.5 H+
(13.15)
The importance of nitric oxide as an intermediate step here was recently established (Caranto and Lancaster, 2017). A portion
of the NADH generated in Equation 13.15 is used to regenerate the NADH used in Equation 13.14, which makes the net yield of
NADH at this point only ½ mol NADH per mole NH+ 4
oxidized.
Nitric oxide is then oxidized to nitrate, capturing more energy in NADH:
NO + 0.5 NAD+ + 0.5 O2 + H2O = NO−

2 + 0.5 NADH
(13.16)
Electron supply and acid generation occur here, when the reaction proceeds all the way through oxidation of HNO toNO−
2 . In
the complete oxidation of ammonium to nitrite, Equations 13.14 to 13.16, one net mole of NADH is generated.
The overall energy-yielding reaction for ammonium oxidation to nitrite is given byEquation 13.17, normalized to 1 e‑ eq:
+
1/6 NH+ −
4 + 1/4 O2 → 1/6 NO2 + 1/3 H + 1/6 H2O
(13.17)
ΔG 0' = -45.44 kJ per e− eq
The value for ΔG 0 ′ is relatively small for aerobic bacteria; in comparison, it is −106 kJ/e− eq for acetate. That, together with the
loss of energy and electrons in ammonium conversion to hydroxylamine, is why AOB have such small values of Y and μ̂ .
The ammonium-oxidizing nitrifiers, all of which have the genus prefixNitroso, are quite genetically diverse. The most commonly
recognized microorganisms that carry out this first step are AOB of the Nitrosomonas genus. Genus names of other AOB able
to oxidize NH+
4
to NO−2 are Nitrosococcus, Nitrosopira, and Nitrosovibrio. The AOA microorganisms are Nitrosolobus,
Nitrosopumilus, Nitrosoarchaeum, and Nitrosotenuis (Herbold et al., 2017). This wide diversity suggests that Nitrosomonas is not
always the dominant genus oxidizing ammonium in any given system. The AOA tend to have a much higher affinity for
ammonium (much lower Ks value) than AOB organisms, but a much lower maximum growth rate (μ̂ ), traits that cause them to
dominate primarily in oligotrophic environments (Martens-Habbena et al., 2009), while the AOB predominate in biological
treatment reactors.
−
The second stage of nitrification is the oxidation of NO−
2 to NO3 (nitratification):
−
1/2 NO−
2 + 1/4 O2 → 1/2 NO3
(13.18)
ΔG 0' = − 38.85 kcal per e− eq
This also is a small ΔG 0 ′ value. Presently, no archaea are known to carry out this reaction. The known genera of nitrite-oxidizing
bacteria (NOB) are Nitrobacter, Nitrospira, Nitrospina, Nitrococcus, and Nitrocystis. The most noted in the past was Nitrobacter,
but molecular probes developed by Mobarry et al. (1996) found that this genus is not the most important in most wastewater-
treatment systems; Nitrospira seems to be more prevalent in treatment processes.
An unexpected discovery came from van Kessel et al. (2015), who found a Nitrospira species that can oxidize ammonium and
nitrite to nitrate, which opens up the potential for treatment by one microorganism that can completely oxidize ammonium to
nitrate; they are termed comammox species. Kits et al. (2017) first isolated a comammox microorganism, naming itNitrospira
inopinata, and indicated it to be slow grower with a high affinity for ammonium, but accompanied with a low maximum
ammonium-oxidation rate. Thus, N. inopinata would tend to dominate primarily under oligotrophic conditions. How important
comammox microorganisms are in wastewater treatment is not yet known.
Summarized in Tables 13.3 and 13.4 are some of the reported basic and derived stoichiometric and kinetic parameters for
ammonium and nitrite oxidizers, respectively, for wastewater and water treatment (Rittmann and Snoeynik, 1984). The first
observation is that fs0 is very low for each group. Compared to the typical fs0 value of 0.6 to 0.7 for aerobic heterotrophs,
nitrifiers transfer relatively few donor electrons to biomass because of the relatively low reaction energy and high energy cost
of synthesis. The low fs0 values translate directly into low Y values. While the numerical value of Y for the ammonium oxidizers
might not appear to be much lower than for aerobic heterotrophs, this apparent closeness is an illusion caused by using
different units in the denominator. In units of g VSS/g OD, Y for the ammonium oxidizers is approximately 0.1, compared to
about 0.45 for heterotrophs. This low yield value reduces the maximum specific growth rates of both organisms to below 1/d at
20°C. With such small values of μ̂ , θ min
x becomes larger—all values are greater than 1 d.
Despite being slow growers, nitrifiers are able to drive effluent NH+
4
or NO−
2 concentrations to very low levels, since S min
values are well below 1 mg N/l. Indeed, an AOA microorganism named Candidatus Nitrosopumilus maritimus strain SCM1 is
reported to have a Ks value as low as 0.0019 mg/l, although accompanied by an equally low μ̂ value of 0.65 d−1. Such
correlations are common with oligotrophic microorganisms in general. In summary, nitrification can be highly efficient, as long
as the SRT is maintained well above θ min
x and sufficient dissolved oxygen is present.
The temperature effects listed in Tables 13.3 and 13.4 are quite important, because nitrification is sometimes considered
impossible in low water temperatures. However, stable nitrification can be maintained at 5°C or lower, as long as the SRT
remains high enough (Haug and McCarty, 1972). At 5°C and with a safety factor of 5, θx would need to be 3.6(5) = 18 d. Another
problem with low-temperature nitrification is that μ̂ becomes quite small, making recovery of nitrification after a washout a very
slow process. Thus, avoiding nitrifier washout due to excess sludge wasting, low DO, or inhibition must be an absolute priority,
particularly for low temperatures.
Commonly recognized today is that nitrification of ammonium to nitrite rather than all the way to nitrate may provide an
advantage for nitrogen removal, as discussed in Section 13.4, where it is called short-cut nitrification. Stopping at nitrite
(nitritification) was indicated by Hellinga et al. (1998) to be more feasible when wastewater temperatures were 15°C and higher
because then the minimum SRT for nitrification was lower. A comparison among [θ min
x ] values in Tables 13.3 and 13.4
lim
indicates that the temperature effect may not be a strong means for selecting for AOB over NOB. However, KN and S minN values
for ammonium oxidation are lower than for AOB at 15°C and above, and this may explain why higher temperatures work better
for short-cut nitrification.
Table 13.3 Basic and Derived Parameter Values at 5 to 25°C for Ammonium Oxidizers Commonly Found in Wastewater
Treatment Reactors
Parameter Variations with Temperature
Parameter 5°C 10°C 15°C 20°C 25°C
fs0 0.14 0.14 0.14 0.14 0.14
Y, mg VSS/mg NH +
4
-N 0.33 0.33 0.33 0.33 0.33
q̂ n, mg NH +
4
-N/mg VSS-d 0.96 1.3 1.7 2.3 3.1
μ̂, d−1 0.32 0.42 0.58 0.76 1.02
KN , mg NH +
4
-N/l 0.18 0.32 0.57 1.0 1.50
KO, mg O 2/l 0.50 0.50 0.50 0.50 0.50
b, d−1 0.045 0.060 0.082 0.11 0.15
[θmin
x ]lim, d 3.6 2.8 2.1 1.5 1.2
SminN , mg NH +
4
-N/l 0.029 0.053 0.096 0.17 0.30
SminO, mg O 2/l 0.081 0.083 0.084 0.085 0.085
Table 13.4 Basic and Derived Parameter Values at 5 to 25°C for Nitrite Oxidizers Commonly Found in Wastewater Treatment
Reactors
Parameter Variations with Temperature
Parameter 5°C 10°C 15°C 20°C 25°C
fs0 0.10 0.10 0.10 0.10 0.10
YN , mg VSS/mg NO−
2 -N 0.083 0.083 0.083 0.083 0.083
q̂ n, mg NO−
2 -N/mg VSS-d 4.1 5.5 7.3 9.8 13.0
μ̂, d−1 0.34 0.45 0.61 0.81 1.1
KN , mg NO−
2 -N/l 0.15 0.30 0.62 1.3 2.7
KO, mg O 2/l 0.68 0.68 0.68 0.68 0.68
b, d−1 0.045 0.060 0.082 0.11 0.15
[θmin
x ]lim, d 3.5 2.6 1.9 1.4 1.1
SminN , mg NO−
2 -N/l 0.024 0.047 0.11 0.20 0.42
SminO, mg O 2/l 0.11 0.11 0.12 0.11 0.11
NH NO−
The following equation is an overall, balanced reaction for the complete oxidation ofNH+
4
to NO−
3 -N by nitrifiers when θx = 15
d. It represents a typical situation for both nitrifier types when they occur together.
NH+
4
+ 1.813 O2 + 0.134 CO2 = 0.0268 C5H7O2N + 0.973 NO−
3
+ 0.946 H2O + 1.973 H+
(13.19)
Besides the low net formation of nitrifier biomass (Y net = 0.22 g VSS/g N; fs = 0.07), this stoichiometric equation illustrates two
other important features of nitrification. First, nitrification creates a major oxygen demand, which is 1.813(32)/14 or 4.14 g O2 /g
NH+ 4
-N consumed. Second, nitrification produces almost 2 strong acid equivalents per mole of NH+ 4
removed. In common
mass units, the alkalinity consumption this represents is 1.973(50)/14 or 7.05 g as CaCO3 per g NH+
4
-N. Without adequate
alkalinity, the resulting pH drop can stop nitrification. The first step, ammonium oxidation, is responsible for the acid production
(Equation 13.17).
Nitrifiers produce soluble microbial products (SMP), which can be consumed by heterotrophic bacteria (de Silva and Rittmann
2000a, 2000b; Rittmann et al., 1994). SMP generally are important in two ways. First, they reduce the net synthesis of the
nitrifiers. Second, they are a way in which nitrifiers can create electron donors for heterotrophs and increase the heterotrophic
biomass.
Nitrifiers are reputed to be highly sensitive to chemical inhibition. This belief is partly accurate. The very slow growth rate of
nitrifiers magnifies the negative impacts of inhibition and, in part, makes it appear that nitrifiers are more sensitive than are
faster growing bacteria. Furthermore, some apparent inhibitors are electron donors whose oxidation depletes the dissolved
oxygen and may cause oxygen limitation, not direct inhibition. However, nitrifiers are sensitive to inhibition from a range of
organic and inorganic compounds. Among the most relevant ones are as follows: unionized NH 3 (at high pH), undissociated
HNO2 (at low pH), anionic surfactants, heavy metals, phenolics, chlorinated organic chemicals, and low pH.
A controversial issue is mixotrophy, in which nitrifiers use an organic carbon source and an inorganic electron donor.
Mixotrophy should significantly reduce the energy cost of cell synthesis, thereby increasing Y and μ̂ . Some claim that almost all
nitrifier strains are capable of mixotrophic growth and have increased values of Y net and μ̂ , but not all research results on this
are consistent. Furthermore, culturing nitrifiers without heterotrophic contamination is extremely difficult. Although mixotrophy
cannot be discounted, when studied carefully, nitrification processes behave in a manner consistent with strictly autotrophic
anabolism.
13.3.2. Common Process Considerations

13.3.2. Common Process Considerations
A successful nitrification process—suspended growth or biofilm—must account for the reality that heterotrophic bacteria
always are present and competing with the nitrifiers for dissolved oxygen and space. The nitrifiers’ relatively high KO values put
them at a disadvantage in the competition for oxygen. Their slow growth rate is a disadvantage when competing for any space
that requires a high growth rate.
These two disadvantages are overcome by ensuring that the nitrifiers are provided a long SRT, typically greater than 15 d,
although larger values may be needed at lower temperatures and in the presence of toxic materials or at low dissolved oxygen
concentrations. In activated sludge, maintaining an SRT of 15 d or greater corresponds to the loading condition termed
extended aeration. Thus, maintaining an extended aeration loading usually is synonymous with having a nitrifying process. For
biofilm processes, the BOD flux and the detachment rate indirectly control the nitrifiers’ SRT.
Even with the relatively long SRT of extended aeration, nitrifying processes often have relatively small safety factors. A low
safety factor is used for economic reasons. When θ min x is 1 to 3 days, the reactor volume can become too great when the safety
factor is greater than about 10, because the θ x/θ ratio cannot be increased indefinitely to compensate. Of course, operating
with a small safety factor increases the risk of washout due to solids loss or inhibition and increases the needs for operator
attention. Unfortunately, the risk is high, and instability in nitrification is a common problem in treatment operations.
13.3.3. Activated Sludge Nitrification: Single-Stage versus Separate-

Stage
Single-stage nitrification is a process configuration in which heterotrophic and nitrifying bacteria coexist in a single reactor with
simultaneous oxidization of ammonium and organic BOD. Figure 13.2a illustrates schematically the single-stage approach.
Single-stage nitrification also can be carried out in sequencing batch reactors (SBRs), which involve sequential periods of filling,
aerobic reaction, settling, and effluent draw-off in one tank. As for any SBR system, multiple units are needed in order that
storage requirements for incoming wastewater are not excessive. Fill-and-draw feeding can create conditions resembling the
more efficient plug flow operation, but with a continuous feed system.
Figure 13.2 Schematic figures of (a) single-stage and (b) separate stage approaches to nitrification
in continuous-flow activated sludge.
Separate-stage nitrification reduces the competition between heterotrophs and nitrifiers by oxidizing most of the organic BOD in
a first stage, while the ammonium is oxidized in a separate second stage. Figure 13.2b illustrates that the separate-stage
process is comprised of two complete activated-sludge systems in series. Because the biomass in each stage is captured and
recycled solely within that stage, each stage develops its own biomass. From an ecological perspective, we have two different
communities, one for each stage. The first stage is essentially free of nitrifiers, while the second stage has a major fraction of
nitrifiers, since all the nitrification but only a small amount of the BOD oxidation results there.
Example
Example 13.4: Single-Stage Design
We are to make a preliminary design of a single-stage nitrification system for a wastewater that has the following
characteristics:
Q = 400 m3/d
BODL° = 500 mg/l
TKN° = 60 mg/l
Inert TKN° = 2.6 mg/l
Inert VSS° = 25 mg/l
T = 15°C
Note that the influent concentration is expressed as total Kjeldahl nitrogen (TKN), instead of NH+
4
-N. Frequently,
wastewaters contain a significant fraction of their reduced nitrogen as organic nitrogen. In any system active in nitrification,
a portion of the organic nitrogen is hydrolyzed to form NH+4
-N, which will be available for nitrification. With the estimate
that 2.6 mg/l TKN° is inert, the total ammonia that needs to be considered for treatment would thus be 60 – 2.6 mg/l or 57.4
mg/l. We also will assume that the inert TKN° is part of the inert VSS°.
First, we must select a design θx. Because [θ min

x ]lim is much larger for nitrification, we use an appropriate value for that
process. Tables 13.3 and 13.4 indicate that [θ min
x ]lim is approximately 2 d for T = 15°C. Applying a safety factor of 10 gives
θ x = 10(2) d = 20 d
Second, we compute the steady-state reactor and effluent concentrations of BODL, NH+
4
-N, and NO−
2 -N from typical
kinetic parameters:
1 + bθ x
S=K
Y q̂ xθ x − (1 + bθ x)
⎡ 1+(
0.1
) (20 d) ⎤
10 mg BODL ⎢ ⎥
) ⎢ ⎥
d
BODL = (
l ⎢
⎢ 0.45 mg VSSa 10 mg BODL ⎥
⎥
0.1
⎣ ( mg BOD ) ( mg VSS − d ) (20 d) − (1 + ( d ) (20 d)) ⎦
L a
= 0.34 mg BODL/l
⎡ 1+(
0.082
) (20 d) ⎤
0.57 mg N ⎢ ⎥
)⎢ ⎥
d
NH4 -N = (
+
l ⎢
⎢ 0.33 mg VSSa 1.7 mg N 0.0.082 ⎥
⎥
⎣ ( mg N
)(
mg VSS − d
) (20 d) − (1 + (
d
) (20 d)) ⎦
a
= 0.18 mg NH+
4
-N/l
⎡ 1+(
0.082
) (20 d) ⎤
⎢ ⎥
)⎢ ⎥
0.62 mg N d
NO–2-N = (
l ⎢
⎢ 0.083 mg VSSa 7.3 mg N 0.0.82 ⎥
⎥
⎣ ( mg N
)(
mg VSS − d
) (20 d) − (1 + (
d
) (20 d)) ⎦
a
= 0.17 mg NO−
2 -N/l
Third, we determine appropriate values of fs and fe for the reactions of interest through use of Equation 13.13.
1 + 0.2(0.1)(20)
Organic oxidation: fs = 0.64 = 0.30,
1 + 0.1(20)
fe = 1 − 0.30 = 0.70
1 + 0.2(0.082)(20)
Ammonium oxidation: fs = 0.14 = 0.070,
1 + 0.082(20)
fe = 1 − 0.07 = 0.93
1 + 0.2(0.082)(20)
Nitrite oxidation: fs = 0.10 = 0.050,
1 + 0.082(20)
fe = 1 − 0.05 = 0.95
Fourth, using Equation 13.4, we determine results from aerobic organic oxidation:
0.70 0.30 0.30

8 g COD + O2 + NH+
4
= C 5H7O2N
4 20 20
(500 − 0.34) mg 400(499.66)

BODL consumed = = 499.66 mg/l or = 200 kg/d
l 1000
499.66 0.70(32)
O2 consumed = [ ] = 350 mg/l or 0.4(350) = 140 kg/d
8 4
499.66 0.30(113)
VSS produced = [ ] = 106 mg/l or 0.4(106) = 42.4 kg/d
8 20
499.66 0.30(14)
NH+
4
-N used for synthesis = [ ] = 13.1 mg/l or 0.4(13.1) = 5.24 kg/d
8 20
Fifth, using Equation 13.6, we determine results from ammonium oxidation to nitrite:
1 1
( + s ) NH+
f f f
4
+ e O2 = NO–2 + s C 5H7O2N
6 20 4 6 20
The quantity of ammonium-N available for consumption equals the influent TKN° minus the inert amount, that left over from
the ammonium reaction, and that consumed by the heterotrophs. Using the above ammonium oxidation equation, the
quantity of other items consumed and produced are readily determined.
Ammonia − nitrogen consumed = (60 − 2.6 − 0.18 − 13.1) = 44.1 mg/l
We also need to leave some small amount of ammonium to be used for cell formation during nitrite oxidation. We have
found that this is only 0.2 mg/l ammonia-nitrogen in the calculations below for nitrite oxidation, and so let’s make that
correction here. That means only 43.9 mg/l ammonia will be available for ammonium oxidation:
0.93(32) 1 0.07
O2 consumed = 43.9[ ]/ [( + ) 14] = 137 mg/l or 0.4(137) = 54.8 kg/d
4 6 20
0.07(113) 1 0.07
VSS produced = 43.9 [ ] / [( + ) 14] = 7.29 mg/l or 0.4(7.29) = 2.92 kg/d
20 6 20
14 1 0.07
NO–2-N produced = 43.9 ( ) / [( + ) 14] = 43.0 mg/l
6 6 20
Sixth, using Equation 13.7, we determine results for nitrite oxidation:
O2 + ( s ) NH+
fe fe f f fs
NO−
2 + 4
= e NO−
3 + C 5H7O2N
2 4 20 2 20
0.95(32) 0.95
O2 consumed = 43.0 [ ] / [( ) 14] = 49.1 mg/l or 0.4(49.1) = 19.6 kg/d
4 2
0.05(113) 0.95
VSS produced = 43.0 [ ] / [( ) 14] = 1.83 mg/l or 0.4(1.83) = 0.73 kg/d
20 2
Check on cell organic nitrogen = 1.83(14/113) = 0.2 mg N/l (check!)
Seventh, we summarize the total reactor O2 mass requirements and VSS mass leaving the reactor:
The total amount of oxygen required = 140 + 54.8 + 19.6 = 214 kg O2 /d
The refractory VSS entering the system = (0.4) 25 = 10.0 kg/d
Total biosolids leaving system = 10.0 + 42.4 + 2.92 + 0.73 = 56.1 kg VSS/d
Eighth, we find the reactor volume. As usual, we must select either a VSS concentration or a hydraulic detention time. In this
case, let’s select Xv = 3000 mg/l. Then,
θ x ΔXv
V = ( )
Xv Δt total
)( ) = 374 m3
20 d 56.1 kg VSS 103 mg ⋅ m3
= (
3000 mg VSS/l d kg ⋅ l
374 m3
θ = = 0.935 d or 22 h
400 m3/d
This is a fairly typical situation for extended aeration (i.e., aθ near one day for domestic wastewater), which almost always
supports single-stage nitrification.
Finally, we provide a preliminary settler design based on the overflow rate and the solids flux. For extended aeration
systems, the overflow rate typically takes a conservative value, 8 to 16 m/d (recall Section 11.7.3). We use 12 m/d. Based
on that criterion, the required surface area for the settler is approximately (we will ignore the small flow rate of waste
sludge):
400 m3/d
As = = 33.3 m2
12 m/d
The solids flux used for extended aeration is also conservative, typically about 70 kg/m2 -d (see Section 11.7.3). To
compute the solids flux, we need an estimate of the recycle flow rate, which requires a judgment about the concentrations
of the underflow solids, Xvr. We assume that Xvr = 8000 mg VSS/l. Then, using the reaction for R = Qr/Q in Example
11.2:
Xv (1 − θ/θ x)
Qr = Q
Xvr − Xv
⎡ 3000 mg/l(1 − 0.935 d ) ⎤
400 m3 ⎢ ⎥
⎢ ⎥ = 229 m3/d (or R = 0.57)
20 d
=
d ⎢
⎢ 8000 mg/l − 3000 mg/l ⎥ ⎥
⎣ ⎦
Having Qr, we can determine the surface area based on the solids flux (Equation 11.26):
(Q + Qr) Xv
As =
Gt
(400 + 229) (3000 mg/l) ( 3 )

m3 10−3kg ⋅ l
d m ⋅ mg
= = 27.0 m2
70 kg/m2 ⋅ d
The overflow rate controls in this case, making As = 33.3 m2 .
Example
Example 13.5: Separate-Stage Design
For the same conditions as in the previous example, we are to make a preliminary design of a separate-stage nitrification
system. At the end, we will compare the single-stage and separate-stage designs.
FIRST STAGE—BODL OXIDATION
In the first stage, we wish to suppress nitrification by using a suitably low θx. Let’s select 4 d. Then, we compute the VSS
production, oxygen required, and unit sizes following the same pattern as in the previous example. (Again, we ignore SMP
for simplicity.) Steady-state concentrations for the first stage become as follows:
1 + 0.1 (4)
BODL = 10 [ ] = 0.84 mg BODL/l
0.45 (10) (4) − (1 + 0.1 (4))
Also here, fs for organic oxidation increases:
1 + 0.2(0.1)(4)
fs = 0.64 = 0.49, fe = 1 − 0.49 = 0.51
1 + 0.1(4)
Again, using Equation 13.4, changes occurring during organic oxidation are determined:
500 − 0.84 0.51(32)

O2 consumed = ( ) = 255 mg/l, or 0.4(255) = 102 kg/d
8 4
500 − 0.84 0.49(113)

VSS produced = [ ] = 173 mg/l or 0.4(173) = 69.2 kg/d
8 20
Total VSS entering or produced in reactor per day = 10.0 + 69.2 = 79.2 kg/d
500 − 0.84 0.49(14)

NH+
4
-N used = [ ] = 21.4 mg/l or 0.4(21.4) = 8.56 kg/d
8 20
One additional calculation we will make, as it is needed for the second stage and overall mass balance, is the concentration
and mass of VSS passing over the weir into the second-stage reactor and the relative proportions of this VSS that are active
and inert.
From the equation for fs, the active proportion of the VSS produced is given by the ratio of 1/[1 + 0.2(0.1)(4)] or 0.93. The
active proportion of the total VSS thus becomes 0.93(69.2)/79.2 = 0.81. Let’s assume that 20 mg/l VSS passes to the
second stage:
VSS passing to the second stage = 20 mg/l or 0.4(20) = 8 kg/d
Next, in order to estimate V and θ, we assume Xv = 3000 mg VSS/l:
)( ) = 106 m3
4d 79.2 kg VSS 103mg ⋅ m3
V = (
3000 mg/l d kg ⋅ l
θ = 106/400 = 0.265 d = 6.4 h
For conventional loading, we can use a settler-overflow rate of 40 m/d. Then, based on the overflow,
As = 400/40 = 10.0 m2
For the solids-flux computation, we need to assume Xvr = 10,000 mg/l and GT = 140 kg/m2-d.
R = 160/400 = 0.4
3000(1 − 0.26/4)
Qr = 400 [ ] = 160 m /d
3
(400 + 160)(3000)(10−3)
10,000 − 3000 As = = 12 m2
140
In this case, the solids flux controls, and As = 12 m2 .
SECOND STAGE—NITRIFICATION
We select the same SRT for nitrification as in the single-stage case, orθx = 20 d. Then, effluent NH+
4
-N and NO−
2 -N remain
at 0.18 and 0.17 mg/l, respectively. The degradable portion of the carryover VSS = 0.81(20 mg/l) = 16.2 mg/l or 0.4(16.2) =
6.5 kg/d. This VSS is partially oxidized in the second stage, releasing some ammonium and lowering the total VSS in the
reactor:
1 + 0.2(0.1)(20)
Influent VSS degraded = 16.2 ( ) = 7.6 mg/l
1 + 0.1(20)
NH+
4
-N released = 7.6(14/113) = 0.94 mg/l or 0.4(0.94) = 0.38 kg/d
Influent VSS remaining = 20 − 7.6 = 12.4 mg/l or 0.4(12.4) = 5.0 kg/d
O2 consumed by VSS decay = 1.42(7.6) = 10.8 mg/l or 0.4(10.8) = 4.3 kg/d
For the AOB, the ammonium-nitrogen available for nitrification must include all inputs and consumptions in the first and
second stage, including the use of about 0.2 mg/l for nitrite oxidizers:
NH+
4
-N available = (60 − 2.6 − 0.18 − 21.4 + 0.94 + 0.2) = 37.0 mg/l
Using Equation 13.6, changes during ammonia oxidation can be determined from
0.93(32) 1 0.07
O2 consumed = 37.0 [ ] / [( + ) 14] = 116 mg/l or 0.4(116) = 46.4 kg/d
4 6 20
0.07(113) 1 0.07
VSS produced = 37.0 [ ] / [( + ) 14] = 6.14 mg/l or 0.4(6.14) = 2.46 kg/d
20 6 20
14 1 0.07
2 -N produced = 37.0 (
NO− ) / [( + ) 14] = 36.2 mg/l
6 6 20
For the NOB, using Equation 13.7 for nitrite oxidation,
0.95(32) 0.95
O2 consumed = 36.2 [ ] / [( ) 14] = 41.4 mg/l or 0.4(40.1) = 16.6 kg/d
4 2
0.05(113) 0.95
VSS produced = 36.2 [ ] / [( ) 14] = 1.54 mg/l or 0.4(1.54) = 0.62 kg/d
20 2
Summarizing the sum total for both reactors of oxygen mass consumption and VSS mass leaving the system:
Second-stage O2 requirement = 4.3 + 46.4 + 16.6 = 67.3 kg/d

Total O2 requirement = 102 + 67.3 = 169 kg/d
Net total VSS production = 79.2− 8.0 + 5.0 + 2.46 + 0.62 = 79.3 kg/d
VSS leaving the second-stage reactor = 5.0 + 2.46 + 0.62 = 8.1 kg/d
To compute the total system volume for the second stage, we must assume a reasonable MLVSS concentration. The low
VSS production generally prevents obtaining the typical values seen when heterotrophic bacteria are dominant. Therefore,
we assume here that Xv = 1000 mg/l. Then,
V = 20(8.1)(103)/1000 = 162 m3
and
θ = 162/400 = 0.405 d or 9.7 h
For the settler, we will use the extended aeration values of overflow rate and solids flux. With the overflow rate of 12 m/d,
400 m3/d
As = = 33.3 m2
12 m/d
The solids flux can be used after assuming an Xvr value of 5000 mg/l.
400(1000)(1 − 0.405/20)
Qr = = 98.0 m3/d
5000 − 1000
and
R = 98.0/400 = 0.25
Then,
(400 + 98.0)(1000)(10−3)
As = = 7.1 m2
70
Again, the overflow rate controls, giving As = 33.3 m2 .
Table 13.5 contains a comparison between the single-stage and separate-stage designs from Examples 13.4 and 13.5. It
demonstrates the general trade-off between the two approaches.
The single-stage system has less biosolids production, since its SRT for BOD oxidation is large, but its oxygen requirement
is greater.
The total system volume is greater for the single-stage system, again due to the long SRT for the heterotrophs, but the
separate-stage system has two separate reactor-settler systems to operate and also has more settler area.
Table 13.5 Comparison Between Single-Stage and Separate-Stage Nitrification Designs from Examples 13.4 and 13.5
Separate Stage
Parameter Single Stage Stage One Stage Two Total
θx, d 20 4 20
MLVSS, mg/l 3000 3000 1000
Solids wasting, kg VSS/d 56.1 71.2 8.1 79.3
Oxygen required, kg O 2/d 214 102 67.3 169
System volume, m 3 374 106 162 268
Settler area, m 2 33.3 10.0 33.3 43.3
13.3.4. Biofilm Nitrification

All of the biofilm processes useful for aerobic BOD oxidation also can be employed to perform nitrification, provided that the
process design appropriately considers the slow specific growth rate inherent with nitrifiers and the competition between
nitrifiers and heterotrophs for oxygen and space. Typical nitrification parameters are listed in Table 13.6.
NH
Table 13.6 Fundamental and Derived Parameters for NH+
4
-N Oxidation
Parameter Value
Notes:
1 Generic values are for T = 15°C.
2 The limiting substrate is NH+
4 -N, which assumes that the first step of nitrification is limiting.
3The b det value assumes that the nitrifiers exist mainly deep within a multispecies biofilm and are partially protected from
detachment.
Limiting substrate NH +
4
-N
Y, mg VSS a/mg NH +
4
-N 0.33
q̂ , mg NH +
4
-N/mg VSSa-d 1.7
K, mg NH +
4
-N/l 0.57
b, d−1 0.08
bdet, d−1 0.05
b′, d−1 0.13
θx, d 20
D, cm2/d 1.3
Df, cm2/d 1.04
Xf, mg VSS a/cm3 10
L, cm 0.004
Sbmin, mg NH +
4
-N/l 0.17
Sb∗min 0.30
K* 1.85
JR, mg NH +
4
-N/cm2-d 0.072
Because BOD almost always is present in wastewaters and because nitrifiers generate SMP that further supports heterotrophic
growth (Rittmann et al., 1994), coexistence with heterotrophs is a normal situation for nitrifying bacteria in all nitrification
processes. Experimental and modeling results indicate that the nitrifying bacteria tend to accumulate closer to the attachment
surface, while heterotrophs dominate near the outer surface of the biofilm (Furumai and Rittmann, 1994; Kissel et al., 1984;
Rittmann and Manem, 1992). This “layering” is a natural consequence of the heterotrophs’ much faster specific growth rate,
which allows them to exist stably in the region of the biofilm experiencing the greatest detachment and predation rates. On the
other hand, the nitrifiers accumulate most successfully deeper inside the biofilm, where they are at least partly protected from
detachment and predation. This phenomenon is discussed in Chapter 7.
While having the nitrifiers deep inside the biofilm helps stabilize the slow-growing nitrifiers against washout, it also increases
their susceptibility to oxygen limitation, since dissolved oxygen must diffuse through the heterotrophic layer before it reaches
the nitrifiers in a normal biofilm reactor. The added mass-transport resistance and oxygen consumption by the heterotrophs
lower biofilm dissolved-oxygen concentrations available for the nitrifiers. When coupled with the relatively high oxygen
sensitivity of nitrifiers (e.g., KO ≈ 0.1 mg/l for many heterotrophs), these low dissolved oxygen concentrations inside the biofilm
can negate the advantages of protection, unless the bulk-liquid DO concentration is maintained high enough. For example,
Furumai and Rittmann (1994) described declines in nitrifier populations and increases in effluent concentrations of NH+ 4
-N
and NO− 2 -N when the bulk DO concentration dropped below approximately 3 mg/l, even though the nitrifiers were well
protected deep inside the biofilm.
In practice, imposing a maximum BOD or COD surface loading, which is in the range of 2 to 6 kg BODL/1000 m2 -d, controls the
competition by heterotrophs. When this design criterion for the organic loading augments a surface-load criterion for the NH+
4
-
N, nitrification performance usually is stable. This makes it possible to maintain a sufficient bulk-liquid DO concentration and
also prevents medium clogging, short-circuiting, sloughing, and/or the need for excessive backwashing caused by too much
heterotrophic growth. The last two items can lead to such large detachment rates that the nitrifiers may never be able to
establish protected layers, and instead are washed out.
Successfully used N and BODL surface loadings for various biofilm processes are summarized inTable 13.7. The ranges of N
and BODL loads are relatively narrow. The BODL loading for nitrification overlaps the BODL loading used for strictly aerobic BOD
oxidation, which explains why biofilm processes used primarily for BOD oxidation often nitrify, as long as the BOD surface
loading is in the lower part of the typical range. Because organic N generally is hydrolyzed to release NH+
4
-N, the N loading
should be in terms of TKN, not just NH+
4
-N.
Table 13.7 Nitrogen and BOD Loadings Used Successfully in Various Nitrifying Biofilm Processes
Process Type N Surface Loading, a kg N/1000 m 2 -d BOD Surface Loading, kg BOD L/1000 m 2 -d
aThe N surface loading should be computed from the TKN concentration, not only the NH+
4 -N concentration, since organic N can be
hydrolyzed to produce NH+4 -N.
Trickling filters 0.5–0.8 <4.4
Rotating biological contactors 0.2–0.6 <6
Biolite granular filters <0.7 <6
Fluidized beds 0.5 Not given
Circulating bed <1 Not given
When the wastewater’s BODL/TKN ratio is too large to allow both surface-loading criteria to be met simultaneously, staged
treatment can be used to reduce the BOD loading to the nitrification process. Two-stage trickling filters are a good example.
The normal staging in RBCs is another example. Normally, nitrification occurs in the later stages of RBCs, after the BOD loading
is reduced in earlier stages. The reduced growth of heterotrophs in nitrifying stages of RBCs allows high-density media to be
used in order to maximize the specific surface area.
Example
Example 13.6: Basic Nitrification Biofilm Design
A wastewater has a volumetric flow of 1000 m3 /d and influent concentrations of 300 mg/l of BODL and 50 mg/l of TKN. We
are to design a one-stage and then a two-stage system that achieves full nitrification.
First, the total mass loading rates of each electron donor are 300 kg BODL/d and 50 kg TKN/d. The 6:1 ratio is typical for
domestic wastewaters.
Second, we find the required surface area of a one-stage system by computing the areas for BODL oxidation and TKN
nitrification. We then use the larger value. For BODL, we use a maximum surface load of 4 kg BODL/1000 m2 -d (see Chapter
12). That gives a biofilm surface area of
A = (300 kg BODL/d)/(4 kg BODL/1000 m2-d) = 7.5(104) m 2
The similar computation for TKN and with a conservative TKN load of 0.5 kg N/1000m2 -d (Table 13.7) gives
A = (50 kg N/d)/(0.5 kg N/1000 m2-d) = 10(105) m 2
We use the larger value, or 10(105 ) m2 .
Second, we do a two-stage design. We begin with BODL oxidation. From Chapter 12, we select a fairly high surface loading
of 7 kg BODL/1000 m2 -d. This gives the first stage area of
A1 = (300 kg BODL/d)/(7 kg BODL/1000 m2-d) = 4.3(104) m 2
We then compute the area for the second, nitrifying stage. Before we compute the area, we must adjust the TKN loading to
account for N used in the first-stage heterotroph synthesis. An exact computation can be achieved following the methods
discussed in Example 13.5. For this simple example, we assume that the TKN load is reduced from 50 to 40 kgN/d. Then,
we assume a less conservation TKN surface load of 0.8 kg N/1000 m 2 -d (see Table 10.5) and compute the second-stage
area.
A2 = (40 kg N/d)/(0.8 kg N/1000 m2-d) = 5(104) m 2
The total surface area for both stages is 9.3(104 ) m2 , which gives a small area savings over the one-stage design. On the
other hand, we must design, build, and operate two separate systems.
13.3.5. Hybrid Processes

Nitrification is one of the applications for which hybrid suspended-growth/biofilm processes are used to increase the
volumetric loading rate. Hybrid processes were broached in Chapter 7. Nitrification is popular because of the widespread need
to oxidize ammonium in wastewaters and because of the strict requirement that the SRT should be high enough to sustain the
slow-growing nitrifiers. Example 13.7 illustrates how the capacity of an overloaded activated sludge process can be expanded
through adding biofilm surface area.
Example
Example 13.7: Hybrid Single-Stage Nitrification
Example 13.4 provided a design for a single-stage nitrification process treating Q = 400 m3 /d, BODL = 500 mg/l, TKN = 60
mg/l, Xi0 = 25 mg/l, and T = 15°C. An SRT of 20 d gave adequate nitrification and BOD removal and required a total system
volume of 374 m3 and an MLVSS of 3000 mg/l. However, we now assume that structural defects forced shutdown of one-
half of the system’s volume. The settler now is severely overloaded when the MLVSS is increased enough to maintain the
same qx. In fact, the practical maximum MLVSS is only about 3000 mg/l, which makes the maximum SRT 10 d. Nitrification
and effluent VSS are unsatisfactory.
To improve the performance, we propose adding biofilm carriers to the 187 m3 of reactor still in operation. We decide to
add carriers that are cubes 3 mm on a side and having a wet density of 1.02 g/cm3 . The outside area per unit cube mass is
0.0196 cm2 /mg. The assumed maximum density for the cubes in the basin is to be 10% by weight, or 100,000 mg/l. Thus,
the biofilm surface area is up to 100,000 mg/l (0.0196 cm2 /mg) = 1960 cm2 /l. The low-density cubes would be kept
circulating within the tank through the bubble aeration used to supply oxygen.
To achieve the same performance as before, the biofilm should have an average SRT near 20 d. For the biofilm, that means
bdet = 0.05/d, which corresponds to the parameters for nitrification in Table 13.6. A mass balance assuming that all
nitrifiers are in the biofilm yields
0 = Q (N 0 − N) − aV J
We can solve this equation to find whatJ needs to be. Some ammonium nitrogen will be used by the heterotrophs. To
account for this and other factors, we will assume the ammonium nitrogen to be consumed by the nitrifiers is the same as
for the single-stage reactor described in Example 13.5, or 44.1 mg/l. We do not yet know the value of N, but it will be small
compared with N0 ; so, let’s assume it to be zero. Then,
44.1 mg NH+ -N 0.0481 mg NH+ -N

)( )(
QN 0 400 m3 l 1
=( )( )
4 4
J= =
aV d l 1960 cm2 187 m3 cm2-d
Using the pseudo-analytical solution for a steady-state biofilm described inChapter 7, we find through iteration the reactor
substrate concentration to be an acceptable 0.5 mg NH4 °-N/l.
For the BOD removal and heterotrophs, it is not necessary that all the biomass should be in the biofilm or that the biofilm
SRT should be 20 d. For this situation, we assume that bdet = 0.1/d for the heterotrophs, and that the parameters of Table
12.1 apply. Using the approach for calculating the flux for a steady-state biofilm fromChapter 7 and the values for BODL
removal in a biofilm from Table 12.1, the flux is calculated to be 0.018 mg/cm2 -d. The concentration of BOD L removed by
the biofilm portion of the reactor would then be
aV J 1960(187)(0.018)
ΔS = = = 16.5 mg/l
Q 400
This is obviously a very small portion of the 500 mg/l BODL entering the reactor; therefore, most of the BODL removal would
have to be performed by suspended heterotrophs in the reactor. The concentration of MLSS would need to be doubled. For
this emergency situation, we might decide to sacrifice some BODL removal and allow the effluent concentration to
increase.
A question to be asked then is what value of J would be required to remove say 50% of the BODL by the biofilm and 50% by
dispersed growth organisms? The answer is
ΔSQ 250(400)
J50 = = = 0.27 mg BODL/cm2-d
aV 1960(187)
Again using the steady-state biofilm model from Chapter 7 and iteration, we find that this much higher BODL flux could be
obtained if we accepted an effluent substrate BODL of 6 mg/l.
This example shows that the treatment capacity of a suspended-growth process could be significantly increased if a
sufficiently large surface area is added and if the effluent substrate concentration is allowed to rise. The success of such a
scheme depends upon having sufficient aeration capacity to keep the biofilm cubes suspended and, most importantly, to
supply oxygen at a volumetric rate approximately twice that of the system discussed in Example 13.4.
13.3.6. The Role of the Input BOD L/TKN Ratio

Nitrifying systems are affected by the influent BODL/TKN ratio in three ways. First, as illustrated in Examples 13.4 and 13.5, is
that synthesis of heterotrophic biomass sequesters nitrogen and reduces the flow of nitrogen from ammonium to nitrite and
nitrate. If the influent BOD L/TKN ratio is large enough—say greater than about 25 g BODL/g TKN—little or no reduced nitrogen is
available for nitrification.
Second, the BODL/TKN ratio determines what fraction of the active biomass is comprised of nitrifiers. Due to the lowfs0 for the
nitrifiers, their fraction normally is low. For a typical BODL/TKN ratio in municipal wastewater (5 to 10 g BODL/g N), the nitrifiers
normally constitute less than 20% of the active biomass and are a smaller fraction of the volatile suspended solids.
Finally, the BODL/TKN ratio exerts some control over how heterotrophs and nitrifiers compete for common resources, dissolved
oxygen and space in flocs or biofilms. In the long term, a higher BODL/TKN ratio tends to force the nitrifiers deeper into the floc
or biofilm. This incurs greater mass-transport resistance for the nitrifiers’ substrates, particularly NH+
4
and O2 (Kissel et al.,
1984; Rittmann and Manem, 1992). In the short term, a high growth rate for heterotrophs could create negative impacts on
nitrifiers by sequestering nitrogen, consuming oxygen, or physically sweeping nitrifiers out of the floc or biofilm.
13.4. Denitrification
Denitrification is the dissimilatory reduction of NO− − − −
3 or NO2 to N2 gas. In other words, NO3 and NO2 are the electron
acceptors used in energy generation. Denitrification is widespread among heterotrophic and autotrophic bacteria, many of
which can shift between oxygen respiration and nitrogen respiration.
In environmental biotechnology, denitrification is applied when the complete removal of N is desired. Key examples include
advanced treatment of wastewater discharged to watersheds that must be protected against eutrophication; treatment of high-
N wastes, such as agricultural run-off and wastewater from feedlots; and nitrogen removal from drinking waters that contain
elevated NO− −
3 + NO2 levels, thereby lowering methemoglobinemia risks to infants. In almost all of these applications, soluble
oxidized forms of N are converted to N2 gas, which evolves to the atmosphere. In order to have denitrification, nitrogen must be
in one of its oxidized forms, NO− −
3 or NO2 . Because many wastewaters contain reduced nitrogen, denitrification frequently is
coupled to nitrification, which is needed to create the oxidized nitrogen forms.
The discussion on denitrification is divided into three sections. Section 13.4.1 contains a review of the fundamentals of
denitrification and denitrifying microorganisms, including the kinetic characteristics of the denitrification reactions. Section
13.4.2 includes descriptions of the various treatment systems commonly used for denitrification.Section 13.4.3 covers the
various biological processes involved in denitrification, including a comparison of the denitrification efficiencies, resources
used and produced, and energy relationships among the different processes.
13.4.1. Physiology of Denitrifying Bacteria
Denitrification is widespread in nature. Denitrifiers are common among the gram-negative Proteobacteria, such as
Pseudomonas, Alcaligenes, Paracoccus, and Thiobacillus. Some gram-positive Bacteria, including Bacillus, can denitrify. Even a
few halophilic Archaea, such as Halobacterium, are able to denitrify. Most of the denitrifiers are facultative aerobes, which
means that they shift to NO− −
3 or NO2 respiration when O2 becomes limiting.
The denitrifiers used in environmental biotechnology are chemotrophs that can use either organic or inorganic electron donors.
Those that utilize organic electron donors are heterotrophs and are widespread among Proteobacteria. A group of autotrophic
denitrifiers can utilize H2 or reduced sulfur as electron donors. A unique group included in the denitrifiers is the anammox
microorganisms, autotrophs that use ammonia as the electron donor, oxidizing it with nitrite and converting both to N2 gas. The
term anammox is short for anaerobic ammonia oxidation, which indeed is descriptive. However, anammox microorganisms
also can be included among the denitrifiers, as they generate N2 by using ammonium as their electron donor; this is why we
consider them in the denitrification section, not in the former section on nitrification. Because of their great metabolic diversity,
denitrifiers of all types are common in soils, sediments, surface waters, groundwaters, and wastewater treatment plants. In
fact, denitrifiers are among the most ubiquitous microorganisms on the planet.
Denitrification proceeds in a stepwise manner in which nitrate (NO− −

3 ) is sequentially reduced to nitrite (NO2 ), nitric oxide
(NO), nitrous oxide (N2 O), and N2 gas. Each half-reaction and the enzyme catalyzing it are as follows:
NO–3 + 2e– + 2H+ = NO–2 + H2O Nitrate reductase

NO–2 + e– + 2H+ = NO + H2O Nitrite reductase
1 1
NO + e– + H+ = N2O + H2O Nitrous oxide reductase
2 2
1 1 1
N2O + e– + H+ = N2 + H2O Nitrous oxide reductase
2 2 2
The overall reaction from NO− −

3 to N2 reduces nitrate by 5 e eq. The first step is a two-electron reduction, while the following
three steps are one-electron reductions.
The dissolved-oxygen concentration controls whether or not microorganisms respire nitrogen. The oxygen concentration can
control denitrification in two ways. The first is repression of the several nitrogen-reductase genes. Research with Pseudomonas
stutzeri indicates that these genes are repressed by dissolved oxygen concentrations greater than 2.5 to 5 mg O2 /l (Körner and
Zumft, 1989). The second control mechanism is inhibition of the activity of the reductase by dissolved-oxygen concentrations
greater than a few tenths mg O2 /l (Rittmann and Langeland, 1985; Tiedje, 1988). The fact that the dissolved-oxygen
concentrations repressing the reductase genes are much higher than the concentrations inhibiting their activity means that
denitrification can occur when dissolved-oxygen concentrations are well above zero (de Silva and Rittmann, 2000a, 2000b;
Rittmann and Langeland, 1985). This situation is enhanced when the denitrifying bacteria are located inside flocs or biofilms,
where the dissolved-oxygen concentration is lower than in the bulk liquid (Rittmann and Langeland, 1985).
Although denitrifiers are not especially pH sensitive, pH values outside of the optimal range of 7 to 8 can lead to accumulation
of intermediates. In low-alkalinity waters, pH control can become a benefit, because denitrification produces strong base. Base
production is illustrated by the balanced reactions in which acetate and H2 are electron donors for the heterotrophic and
autotrophic denitrifiers, respectively.
8 4 4 8
CH 3COOH + NO–3 + H2O = N2 + 2H2CO3 + OH–
5 5 5 5
(13.19)
8 4 8 16
4H2 + NO–3 = N2 + OH− + H2O
5 5 5 5
(13.20)
In both cases, the water’s alkalinity is increased by the 8/5 equivalents of strong base produced when 8/5 mol ofNO− 3 -N is
−
reduced. In mass terms, that is an alkalinity increase of (50)/(14) = 3.57 mg alkalinity as CaCO 3 /mg NO3 -N consumed. For the
acetate case, this effect is altered somewhat because 2 mol of a weaker acid (H2 CO3 , pKa ≈ 6.3) replace 1 mol of a weak acid
(CH3 COOH, pKa ≈ 4.3).
Although the organic substrates used by heterotrophic denitrifiers cover a nearly infinite range, only a few simple and
inexpensive organic substrates have generally been considered for use in practice. One of the first such uses was for
denitrification of agricultural runoff waters that contained high nitrate concentrations, but no organics (McCarty et al., 1969).
Methanol (CH3 OH), which was among those evaluated (methanol, acetate, glucose, and ethanol), was found effective and
relatively inexpensive; mainly because of its low cost, methanol has gained frequent use as an organic electron donor for
denitrification. However, the bacteria that utilize methanol are methanotrophs, which have some special biochemical
characteristics as one-carbon oxidizers.
Table 13.8 contains representative stoichiometric and kinetic parameters (at 20°C) for microorganisms using methanol,
wastewater BOD, H2 , and elemental sulfur (S), and ammonium as the electron donor for denitrification. The values in Table 13.8
lead to the following observations about denitrifiers and how processes involving denitrification should perform.
1. While the fs0 value for heterotrophs using general BOD is only slightly smaller than fs0 for aerobic heterotrophs (around 0.6),
the fs0 values for the autotrophic denitrifiers are much smaller, similar to that for nitrifiers.
2. The fs0 value for the one-carbon oxidizers that consume methanol is lower than for other heterotrophs.
3. True yield values parallel the fs0 values.
4. Since q̂ and b values are roughly similar, about 12 g O2 /g VSS a-d (not shown in Table 13.8) and 0.05/d, [θ min
x ]lim is
0
controlled mainly by fs or Y.
5. S min values are less than 1 mg/l, which means that one should be able to prevent high donor residuals in the effluent.
∗ ) when detachment is negligible. On the other
6. For biofilm processes, all microbial types show high growth potential (low Smin
∗ ).
hand, the autotrophic processes are subject to growth limitations as bdet increases (higher Smin
Table 13.8 Representative Stoichiometric and Kinetic Parameters for Denitrifiers, T = 20°C
Electron Donor Methanol BOD H2 S° NH +

4
-N (anammox)
For K*, L = 40 µm, D f/D = 0.8, and Xf = 40 mg VSSa/cm3; n.a., not applicable; n.f., not feasible.
C—source Methanol Organics CO 2 CO 2 CO 2
fs0 0.36 0.52 0.21 0.13 0.14
Y, g VSS a/g donor 0.27 0.26 0.85 0.10 0.105
q̂ , g donor/g VSSa-d 6.9 12 1.6 8.1 0.73
K, mg donor/l 9.1 1 1 n.a. 1
b, d−1 0.05 0.05 0.05 0.05 0.02
[θmin
x ], d
0.55 0.33 0.76 1.3 10
Smin, mg donor/l 0.25 0.017 0.04 n.a. 0.04
D, cm2/d 1.3 1 1.6 n.a. 1.3
∗ (no detachment)
Smin 0.027 0.017 0.04 0.066 0.04
(b det = 0.2/d)
0.15 0.087 0.23 0.45 n.f.
K* 1.8 0.4 2.9 n.a. 1.9
13.4.2. Denitrification Systems

Figure 13.3 illustrates four different biological treatment systems typically used for nitrogen removal. They are represented in
the following denitrification system comparisons and based upon resources used and produced, potential greenhouse gas
emissions, state of development, and the current potential limitations of each. System A is used for traditional organic
oxidation along with nitrification/denitrification using an external organic electron donor for denitrification. Here, wastewater
organics are oxidized in a first reactor, leaving none for denitrification later on. Nitrification is carried out along with organic
oxidization in the first reactor or possibly in a second reactor. Since no organics then remain, a donor must be added to the final
reactor for denitrification. Methanol was found to be effective for this purpose, as it generally is the least expensive among
other simple organics that may be used, such as acetate, ethanol, and sugar (McCarty et al., 1969). Methanol is still used at
some treatment plants.
Figure 13.3 Schematics of nitrogen-removal systems. (Source: Adapted with permission from
McCarty (2018), copyright 2018 American Chemical Society.)
System B is a proposed system that might be used to carry out denitrification with anammox microorganisms, although
efficient full-scale application has not yet been demonstrated. System B is well worth considering as we move toward better
environmental sustainability. Again, organics are removed in the first reactor either aerobically or anaerobically. In the second,
nitrification of about half of the ammonia occurs, but only to nitrite, which is then used in the third reactor as the electron
acceptor for anammox bacteria oxidizing the ammonium and reducing the nitrite formed to N2 . It also is possible that partial
nitrification by nitrifiers and denitrification by anammox organisms can be carried out in a single reactor. The great advantage
over System A is that no external electron donor is needed; indeed, the total oxygen requirement for nitrification is less than
one-half of that needed for System A. If one used anaerobic rather than aerobic treatment for the first reactor of System B, the
oxygen requirement would be further reduced, while producing more energy in the form of methane. The development of an
efficient and reliable system for anaerobic organic oxidation followed by anammox denitrification is being explored by many,
but such a full-scale system is not yet ready for full-scale implementation. The potential energy benefits of doing so could be
substantial.
System C represents a common approach now in widespread use to bring about traditional denitrification without the need for
an external electron donor. This is pre-denitrification by the Modified Ludzack–Ettinger (MLE) process. Here, denitrification is
conducted in the first or anoxic reactor, which is why it is called pre-denitrification. The nitrate or nitrite used for denitrification
is formed in the second reactor and recycled back to the first, which also receives the wastewater organics from the primary
tank so that denitrification of the nitrates and/or nitrites to N2 gas can occur in the first tank. Efficient denitrification here
requires a high recycle rate to minimize the nitrate lost in the effluent. However, recycle from the aerobic second tank also
brings back dissolved oxygen, reducing the efficiency of denitrification. A rate of recycle equal to about five times the influent
flow rate was determined to be about the maximum that should be used (Ekama and Wentzel, 2008). An additional advantage
of this system over System A is that the total O 2 requirement is significantly reduced, since much of the organic material is
oxidized with the nitrate, rather than with oxygen. Indeed, if only nitrification to nitrate was required—in order to reduce oxygen
demand in a receiving water—using System C, with denitrification instead of only nitrification, would accomplish total-N removal
with a smaller needed supply O2 .
System D is one whose application is growing, especially when the BOD/N ratio is too low to carryout complete denitrification,
say by using System C. Here, the side stream from anaerobic digestion of sludge is denitrified using the anammox process.
Advantages over mainstream anammox treatment is that the temperature is high and more favorable for the slow-growing
anammox microorganisms, the diurnal variation in ammonia concentration is smaller than in the mainstream (so that just the
right amount of oxygen addition for partial nitrification can be accomplished), and the waste stream is smaller in flow rate (but
with high ammonium concentration), allowing a side-stream reactor to be relatively small in size.
Not specified here is the type of reactor to be used in the above systems. Either dispersed growth or biofilm reactors can be
employed in any of the systems. They may be completely mixed, plug-flow, or sequential batch systems. Each has advantages
and disadvantages that have been discussed in earlier chapters, and using a mix of reactor types may be advantageous. One
may use primary treatment, as indicated here, or not, depending upon wastewater or other characteristics. Rather than getting
into such details, our objective here is to conduct a relative comparison of system treatment efficiencies, resource needs,
energy requirements or production, effluent characteristics, and the potential for greenhouse-gas production for each system.
Results are presented, compared, and discussed in Section 13.4.3.
13.4.3. Comparing the Nitrogen-Removal Systems

Resources used and produced in the nitrogen removal systems illustrated in Figure 13.3 were evaluated by McCarty (2018).
Used for this comparison was a mass-balance approach based on the skeleton stoichiometric equations (Equations 13.4
through 13.9). Assumed was a base-case municipal wastewater with an influent BODL of 400 mg/l, ammonium and organic
nitrogen concentrations of 30 and 20 mg/l, respectively, and a total VSS concentration of 200 mg/l. The assumed wastewater
flow rate was 4000 m3 /d. Two cases of VSS-removal efficiency for primary settling were used: 65% for typical performance and
100% with enhanced settling from chemical treatment. The biodegradable fraction of the input VSS was assumed to be 76%.
Based on a COD/VSS ratio of 1.42 g COD/g VSS, BOD L removal by primary settling with 100% VSS removal would be 0.76(200)
(1.42)/400 or 54%. If the typical VSS by primary settling were 65%, then BODL removal would be 0.65(54) or 35%. Evaluated
were conventional nitrification of ammonium to nitrate followed by denitrification, as well as short-cut nitrification only to
nitrite. Other assumptions used for these analyses are listed in Table 13.9.
Table 13.9 Assumed Values of fs0, SRT, and Treatment Efficiency Used for Nitrogen Removal Analyses
Reaction fs b d−1 SRT d fs Net Treatment Efficiency %
Source: Adapted with permission from McCarty (2018), copyright 2018 American Chemical Society.
Aerobic organic oxidation
B1 system 0.555 0.05 6 0.449 95
A, C, and D systems 0.555 0.05 12 0.385 95
Organic methanogenesis 0.080 0.03 70 0.037 95
Ammonia oxidation to nitrite 0.090 0.05 12 0.063 90
Nitrite oxidation to nitrate 0.070 0.05 12 0.049 90
Ammonia oxidation to nitrate 0.080 0.05 12 0.056 90
Denitrification of nitrite or nitrate to N 2
A system 0.500 0.05 6 0.408 95
C and D systems 0.500 0.05 12 0.350 95
Anammox 0.080 0.05 20 0.0480 95
Results of the stoichiometric analyses are listed in Table 13.10. Included is the ratio of the energy that could be obtained from
combustion of the methane produced, normalized to the energy required for supplying oxygen (i.e., the CH4 /O2 energy ratio).
Values used were 9.92 kWh/m3 (STP) for methane combustion (Kim et al., 2011) and 1.00 kWh/kg for oxygen usage (Owen,
1982).
Table 13.10 Stoichiometric Analyses for Nitrogen Removal Treatment Systems
N Added O2 Digested CH4 CH4 /O2 Effluent N—mg/l

Removal COD Used Biosolids m 3 /d
Sys. Treatment % kg/d kg/d kg/d STP ER Org. NH 3 NO−

2 NO−
3
A. 65% Primary Suspended Solids Removal
Direct Line Nitrification/Denitrification
A1 Nitratification 98 724 1,272 468 297 2.3 0.9 0.0 0.0 0.0
A2 Nitritification 94 420 1,112 424 280 2.5 0.7 0.6 1.8 0.0
Mainstream Anammox
B1 Aerobic organic 93 832 388 258 3.1 0.6 0.4 1.0 1.4
oxidation
N Added O Digested CH CH /O Effluent N—mg/l
Removal COD Used Biosolids m /d
Sys. Treatment % kg/d kg/d kg/d STP ER Org. NH
B2 Methanogenesis 93 320 252 526 16.3 0.2 0.5 1.1 1.5
Recycle Nitrification/Denitrification
C1 Nitratification 85 960 372 252 2.6 0.1 0.7 0.0 6.8
C2 Nitritification 85 928 372 252 2.7 0.1 0.7 6.8 0.0
Recycle Nitrification/Denitrification Plus Side-Stream Anammox
D1 Nitratification 90 936 372 252 2.7 0.1 0.5 0.0 4.6
D2 Nitritification 90 916 372 252 2.7 0.1 0.5 4.5 0.0
B. 100% Primary Suspended Solids Removal
Direct Line Nitrification/Denitrification
A1 Nitratification 99 744 1,120 452 375 3.3 0.0 0.0 0.0 0.0
A2 Nitritification 94 432 936 408 358 3.8 0.3 0.6 1.9 0.0
Mainstream Anammox
B1 Aer. org. oxid. 93 672 364 336 5.0 0.4 0.6 1.0 1.4
B2 Methanogenesis 93 308 268 526 16.9 0.1 0.7 1.1 1.5
C1 Nitratification 85 784 352 330 4.2 0.1 0.8 0.0 6.9
C2 Nitritification 85 752 352 330 4.4 0.1 0.8 6.9 0.0
D1 Nitratification 91 752 352 330 4.4 0.1 0.4 0.0 4.1
D2 Nitritification 91 732 352 330 4.5 0.1 0.4 4.0 0.0
By far the most environmentally unfriendly was System A1, to which organic COD was added as needed for traditional
nitrification followed by denitrification. The oxygen usage and digested biosolids production are by far the highest of any of the
other systems, and the low CH4 /O2 ratio is the poorest as well. System A2, with nitrification only to nitrite, is somewhat more
favorable, but still inferior to other systems. The sole advantage of System A is a slightly higher nitrogen-removal percentage.
The best system by far is System B2, mainstream anaerobic treatment followed by anammox nitrogen removal. The need for O2
and digested biosolids production is far below that of any other system, while methane production and the CH4 /O2 energy ratio
are far greater. The challenge here is that efficient full-scale implementation has not yet (2020) been achieved for mainstream
anaerobic treatment or for mainstream anammox treatment. However, both have been demonstrated at lab and pilot scales,
and considerable ongoing research is directed toward making their general use possible because of the significant potential
advantages offered.
Another important comparison is with System B1, aerobic mainstream treatment followed by anammox nitrogen removal. This
potential combination often is claimed to have a significant advantage over traditional nitrification/denitrification. The
advantage certainly is evident in a comparison with Systems A1 and A2, but not when compared with Systems C1 and C2. The
latter, which use anaerobic before aerobic treatment, take advantage of using the oxidized nitrogen forms as electron acceptors
rather than O2 for organic oxidation. Thus, added organics are not needed as long as the ratio of BOD-to-ammonium-nitrogen of
the wastewater is high enough. System B1 has about 15% less usage of O2 than with the C recycle systems, and B1 also has a
somewhat higher CH4 /O2 energy ratio; however, biosolids and methane productions are similar.
Mainstream anammox treatment is more complicated than traditional nitrification/denitrification, and this currently is holding
up its use for mainstream applications. The anammox microorganisms are exceptionally slow growers and require a long SRT
in order to be sufficiently active. Additionally, oxidizing just the right amount of ammonium to nitrite for this process, particularly
supplying just the right amount of O2 to meet a widely variable diurnal inlet ammonium concentration, is challenging. If too
much O 2 is provided, oxidation of ammonium all the way to nitrate by traditional nitrifiers occurs. If too little O2 is supplied,
oxidation of ammonium to nitrite is not sufficient. Either way, nitrogen removal will be insufficient. Thus, using a recycle system
with traditional nitrification/denitrification appears much easier and more reliable than mainstream annamox, although
requiring somewhat more oxygen and energy.
The use of System D is growing today, especially if the BOD/NH+

4
-N ratio is low. Anammox treatment is used for the side-
stream denitrification: i.e., for removing the higher ammonium-nitrogen concentration in digester supernatant liquid. Here, the
temperature is higher and more favorable for anammox microorganism growth. The main advantages, illustrated in Table
13.10 for System D over System C, are that N removal is about 5% higher and O2 used is only about 2% lower; these are not
strong reasons to add a complex component to a treatment system unless the influent BOD/NH+ 4
-N ratio is too low for the
more conventional recycle approach.
Another way to reduce oxygen use and increase methane production without using mainstream anaerobic treatment is to
concentrate more degradable organics into the biosolids sent for anaerobic digestion. One approach to accomplish this is
through enhanced chemical precipitation of influent suspended solids prior to primary sedimentation. Another is to use a very
short SRT in a first-stage activated sludge process to increase the net yield (that is by increasing fs), which also reduces oxygen
consumption. This change produces results similar to chemically enhanced primary settling. This is illustrated in the lower
portion of Table 13.10. Enhanced removal of BOD in primary sedimentation significantly reduces the BOD available for
traditional nitrification/denitrification, but, for the influent concentrations assumed for this study, the available BOD is still
sufficient to avoid a BOD deficiency for denitrification.
The relative differences among the systems with 100% primary solids removal do not change significantly over those with 65%
removal. In absolute terms, however, sending a higher percentage of organic BOD to anaerobic digestion provides significant
reductions in O2 requirements and digested biosolids production, along with increased production of methane and the CH4 /O2
energy ratio. Nitrogen removal changes little. While the improvements from System D are significant (compared to Systems A
and C) and are being implemented today, they do not equal the potential advantages coming from System B2, mainstream
anaerobic treatment followed by mainstream anammox treatment. Successful development of an efficient and reliable full-
scale System B2 would be a large step toward better environmental sustainability.
An especially important factor in Table 13.10 is the CH4 /O2 energy ratio. The energy efficiency of converting methane energy
into electrical energy is roughly 33%, which means that an energy ratio of about 3 is needed to offset the energy required for
aeration. As the energy required to supply oxygen is about one-half that needed for overall plant operation, a ratio of 6 would be
needed overall. Enhanced primary settling provides an energy value of 3 or higher in all cases, but only mainstream anammox
treatment would approach the value of 6 with aerobic organic oxidation System B1. However, System B2—mainstream
anaerobic treatment—has an energy ratio greater than 16. Thus, System B2 would be an exceptional net energy producer.
Another aspect of energy production with methane should not be ignored: putting to good use the waste heat produced by
methane combustion. One scheme is to use the waste heat for drying raw or digested biosolids for efficient incineration,
therefore reducing the problem and cost of disposing of waste biosolids and potentially producing even more energy during
combustion (Scherson and Criddle, 2014).
Another important aspect of this analysis is understanding when the influent BOD/NH+
4
-N ratio is too low to bring about
efficient nitrification/denitrification with recycle (Systems C1 and C2). Daigger (2014) determined the theoretical minimum
required BODL/N ratio to be 3.4 to 4.0 with nitratification, 2.0 to 2.5 with nitritification, and 0.5 with anammox. These ranges
reflect the differences in nitrogen used for cell synthesis with the typical SRT ranges used. In the example case for the analyses
in Table 13.10, the incoming ratio of BODL to total N is 400/50 or 8.0 g BODL/g N, much higher than the theoretical needs.
However, some of the incoming N is in refractory organic compounds, and some is removed by primary settling. So what ratio
of BOD L to total N becomes limiting for the overall System C cases evaluated here? In order to examine this, the influent total N
was held at 50 mg/l, while the BODL of 400 mg/l was reduced in steps until the BODL became insufficient (McCarty, 2018). The
influent VSS/BODL ratio was maintained at 200/400 = 0.5 g VSS/g BODL, and the influent org-N/VSS ratio was maintained at 0.1
g N/g VSS. This means that, as influent BODL was lowered, the concentration of ammonium-nitrogen increased to maintain the
total N at 50 mg N/l. The results are summarized in Table 13.11.
Table 13.11 For Systems C and D, the Minimum BODL/N Ratios That Would Permit Good N Removal with 65% and 100%
Primary VSS Removal Efficiencies, 50 mg/l of Influent Total N Concentration, and a Recycle Ratio of 5:1
65% Primary Settling 100% Primary Settling
System Treatment N Removal Minimum BOD L Influent BOD L/N N Removal Minimum BOD L Influent BOD L/N
% mg/l Ratio % mg/l Ratio
C1 Nitratification 84 258 5.2 84 326 6.5
C2 Nitritification 83 164 3.3 83 208 4.2
D1 Nitratification 86 222 4.4 87 260 5.2
D2 Nitritification 85 148 3.0 86 179 3.6
Noteworthy is that nitritification needs a much lower BOD L/N ratio for recycle nitrification/denitrification than does
nitratification. Also, a significantly lower ratio is possible when side-stream anammox treatment is employed, which is a main
reason for its use. The minimum ratios are also higher with 100% primary settling than with 65%, since much of the BODL in the
former is converted into methane and thus not available for denitrification. The practical operational ratios here are 5.2 and 3.3
for nitratification and nitritification, respectively, with normal 65% primary VSS settling and without side-stream anammox
treatment. These ratios are 25% to 40% higher than the theoretical values (Daigger, 2014), primarily because of the BOD
removed with primary settling. Greater removal of BODL by enhanced primary settling can be offset if needed by side-stream
anammox treatment.
In summary, the best choice of which to use of the various processes illustrated inFigure 13.3 depends upon the
characteristics of the wastewater and the particular situation involved. A good understanding of the various factors involved
when designing a wastewater treatment system for nitrogen removal is thus of importance.
13.5. Range of Nitrification and Denitrification Systems

A wide variety of dispersed-growth and biofilm reactors can be used for nitrification and denitrification. The kinetics of the
biological reactions in different reactor types can differ as indicated in Chapters 6, 7, and 9. For a given temperature and SRT,
the factors affecting the efficiency of a given reaction may differ as well. However, for the steady-state case in any reactor with
a given SRT and microorganism decay rate, b, the relative fractions of electron donor converted for energy and for synthesis will
be the same, as indicated by Equation 6.37. This means that, for a given expected treatment efficiency, SRT, andb for design,
we can calculate through stoichiometry (Chapter 5) the quantitative requirements for the electron acceptor and N and P
nutrients, as well as the reaction end products, such as mass of bacterial cells and other end products produced. This has been
illustrated for several nitrogen-removal systems in Section 13.4.3.
Treatment efficiencies of 90% to 95% for various substrates, as used in Section 13.4.3, are typical. The need then when
designing such a reactor type is to select its size and other features so that it actually can meet the SRT and the efficiency
desired. A more detailed examination of reactor kinetics may or may not be such an important part of that selection as
indicated in Section 9.2.2. This is especially so for dispersed-growth treatment systems operated with an SRT safety factor of 3
to 6 and an S 0 /K ratio greater than 10. Then, predicted treatment efficiencies for individual substrates over 90% can be
expected. However, that may not be true for biofilm reactors, which have inherent mass-transport resistances. Additional
discussion of mass-transport effects is provided in Section 13.5.1. Section 13.5.4 covers a discussion of side-stream
anammox treatment.
13.5.1. Biofilm Reactors

Denitrification is one of the easiest applications for a wide range of biofilm processes. Eliminating the need for O2 transfer
relieves the biggest limitation on volumetric loading. In addition, accumulation by biofilm attachment eliminates volume
constraints imposed by a settler. Thus, volumetric loading can be much higher than for aerobic biofilm processes.
Although the oxidized form of nitrogen, such as nitrate, is the target pollutant for removal, the electron donor almost always is
the rate-limiting substrate. Therefore, design of biofilm systems for denitrification should be based on the flux of the donor.
Successful operation of a range of different biofilm systems has shown that a donor flux in the high-load region gives
essentially complete N removal in wastewater effluents. In particular, J (flux values) for organic donors may range from 15 to
22 g BODL/m2 -d. Using H2 as an autotrophic electron donor has allowed nearly complete NH+ 4
removal for fluxes up to 15 kg
OD/1000 m 2 -d. From stoichiometry, NO− 2
3 fluxes (g N/m -d) are one-third to one-fourth the value of the necessary donor fluxes.
Almost any biofilm system works well for traditional denitrification, as long as oxygen introduction can be kept at a minimal
level and plugging does not occur. Successful systems include the following:
Rotating biological contactors in which air ventilation is restricted
Submerged fixed beds of rocks, sand, limestone, or plastic media
Fluidized beds of sand, activated carbon, pellets of ion-exchange resin, or naturally formed granules of microorganisms
Circulating beds using a range of lightweight plastic materials
Membrane biofilm reactors with gas-permeable membranes used to supply H2 (or O2 in the anammox case), as well as to
serve as the biofilm attachment surface
To be included are anammox systems that combine partial oxidation of ammonium to nitrite followed by its conversion to N2 ,
while being supplied with just the needed amount of O2 to oxidize about half of the incoming ammonium to nitrite. One
interesting approach is the use of a SBR that is switched back and forth between an aerobic stage to oxidize organics and
ammonium to nitrite and then to an anoxic stage for anammox conversion of ammonia and nitrite to N2 (Lotti et al., 2015;
Winkler et al., 2012). With any biofilm reactor, hydraulic detention times can be low when support media of high specific surface
area are utilized in a non-clogging configuration. For example, fluidized beds and membrane biofilm reactors can give hydraulic
detention times in some cases of less than 10 to 20 min (Rittmann, 2018; Shin et al., 2012; Ziv-El and Rittmann, 2009a).
Example
Example 13.8: Fluidized-Bed Denitrification Using H2
A wastewater containing 50 mg NO− 3

3 N/l in a flow of 1000 m /d is to be treated with autotrophic denitrification in a
fluidized bed. The biofilm detachment rate is controlled such that the average SRT is 15 d. No O2 transfer occurs into the
system. We provide a preliminary design for a fluidized bed in which we use 0.75-mm particles and a bed expansion that
makes the specific surface area 4000 m2 /m3 of bed. We select a H2 flux equivalent to 20 g OD/m2 -d, which gives a highly
loaded process.
Using Equation 6.37 and values from Table 13.8, we find fs and fe:
1 + 0.2(0.05)(15)
fs = 0.21 = 0.138, and fe = 1 − 0.138 = 0.862
1 + 0.05(15)
With Table 5.5, Equation I-5 for H2 as an electron donor and Equation I-7 for nitrate as the electron acceptor, we derive the
following half-equations for the overall reaction:
1 6 + 1 3
Nitrate as acceptor: 0.862 ( NO− −
3 + H +e = N2 + H2O)
5 5 10 5
Synthesis with NH+
4
as N source:
1 1 1 1 9
+ 0.138 ( CO2 + NH+
4
+ HCO− + −
3 +H +e = C 5H7O2N + H2O)
5 20 20 20 20
1
H2 as the donor: + H2 = H+ + e–
2
We can see that the amount of H2 required is
]( 3 ) = 20.7 kg H2/d
50 mg NO–3 1000 m3 0.5(2) g H2 kg-l
H2 = ( )[ –
l d (0.862/5) (14) g NO3-N 10 mg-m3
This converts to
20.7 kg H2 8 g OD
[ ] = 166 kg OD/d
d 0.5(2) g H2
With a specific surface area of 4000 m2 /m3 and a surface loading of 20 g OD/m2 -d = 20 kg OD/1000 m2 -d, the volume
becomes
166 kg OD 1000 m2 m3
V= ( )( ) = 2.08 m3
d 20 kg OD 4000 m2
Then, the hydraulic detention time is
V 2.08 m3
θ calc = = = 0.00208 d = 3.0 min
Q 1000 m3/d
The short detention time could result in a fairly high upward flow velocity, depending upon the cross-sectional area
selected. The upward flow will need to be consistent with the settling velocity of the particles selected for the fluidized bed.
The H2 delivery rate is 20.7 kg H2 /d, which represents a concentration in the 1000 m3 /d of flow of about 21 mg/l, or more
than 10 times the H2 solubility. Thus, a H2 -transfer process will be required and may affect the necessary reactor detention
time. In principle, one could sparge H2 gas into the fluidized bed reactor, but this could create a safety problem due to off-
gassing of the combustible gas. That leads to the next example, in which H2 gas is delivered by diffusion through the wall of
gas-transfer membranes (Lee and Rittmann, 2002).
Example
Example 13.9: H2-Based Denitrification Using Bubbleless Gas Transfer in a Membrane Biofilm Reactor
The same wastewater from Example 13.6 (50 mg NO− 3

3 N/l in a flow of 1000 m /d) is to be treated with autotrophic
denitrification in a membrane biofilm reactor (MBfR). In an MBfR, H2 gas is provided to the lumen of nonporous hollow-fiber
membranes, and it diffuses through the membrane wall and directly into a biofilm that lives on the membrane’s outer
surface. H2 delivery is on-demand in proportion to the need to reduceNO− 3 . The H2 delivery capacity is proportional to the
H2 pressure inside the membranes (Tang et al., 2012). A reasonableNO− −
3 -N flux is 3 g NO3 -N/m -d (Ziv-El and Rittmann,
2
2 3
2009b), and biofilm specific surface area in a commercial-scale MBfR is around 800 m /m . Biofilm detachment rates are
low in MBfRs, and we will control it such that the average SRT is 15 d (as in the previous example). No O2 transfer occurs
into the system.
The stoichiometry is the same as in the previous example, and this means that the H2 demand is 166 kg H2 /d for a total N
removal of 50 kg N/d.
With a specific surface area of 800 m2 /m3 and a surface loading of 3 g N/m2 -d = 3 kg N/1000 m2 -d, the volume becomes
50 kg N 1000 m2 m3
V= ( )( ) = 20 m3
d 3 kg OD 800 m2
Then, the hydraulic detention time is
V 20 m3
θ calc = = = 0.02 d = 30 min
Q 1000 m3/d
Again, this is a short hydraulic retention time. Furthermore, the MBfR eliminates the problem of delivering low-solubility H2
gas.
13.5.2. The Barnard Process for Nitrogen Removal

One operational difficulty with the nitrification/denitrification System C inFigure 13.3 is that the efficiency of nitrate removal is
governed by the recycle rate of fluid from the final reactor and settling tank to the head of the first reactor. In order to achieve
the 85% nitrogen removal listed in Table 13.10 (part A) for System C1, a recycle rate of 5 was used. Recycling fluid requires
energy, and a high recycle ratio also causes other unwarranted problems, particularly too much carryback of dissolved oxygen
from the aerobic system. The Barnard process, developed many years ago by Dr. James Barnard (Barnard, 1975), offers a way
to obtain better removal of nitrogen without increasing the recycle ratio. Shown schematically in Figure 13.4, the Barnard
process begins with classical anoxic reactor for denitrification (Reactor 1) followed by an aerobic reactor for organic oxidation
and nitrification (Reactor 2). The typical hydraulic retention time for these two reactors, based on the plant flow Q, is around 14
h, with a range from 7 to 24 h. With a recycle ratio for QR2 of 400%, approximately 80% of the influent TKN is denitrified to N2 ,
while about 20% leaves reactor 2 as NO− −
3 -N. If the influent TKN were 50 mg/l, the NO3 -N concentration leaving Reactor 2
would be about 10 mg/l.
Figure 13.4 Schematic of the Barnard process.
The effluent from Reactor 2 flows to Reactor 3, where endogenous respiration of biosolids from Reactor 2 fuels denitrification
of the remaining 10 mg NO− 3 -N/l to N 2 gas. Typical hydraulic detention time here is 3 h. Cell decay releasesNH4 -N to the
+
water with a ratio of roughly 0.3 mg NH+

4
-N/mg NO−
3 -N. Thus, the effluent from the biosolids decayer (Reactor 3) contains
about 3 mg NH+
4
-N/l.
Reactor 4 provides about 1 h of aeration in order to oxidize theNH+

4
-N to NO−
3 -N. Hence, the effluent from the Barnard
process contains NO− 3 -N at about 3 mg/l, or 6% of the influent TKN; thus, 94% removal of influent total N is obtained. The
settler performs its usual functions and is designed using extended-aeration criteria.
The Barnard process is well established worldwide as a means to achieve greater than 90% total N removal. Its main liabilities
are the many tanks, the relatively long hydraulic detention time, and the still significant mixed-liquor recycle between Reactors 2
and 1. An added advantage, however, is that phosphorus removal is also obtained in this system, as described in Chapter 14.
13.5.3. Sequencing Batch Reactor

13.5.3. Sequencing Batch Reactor
The sequencing batch reactor (SBR) is seeing greater use for nitrogen removal in mainstream and side-stream applications. An
SBR can get around some of the difficulties with high recycle rates and energy costs of the recycle systems. Figure 13.5
illustrates a typical 9-h cycle for achieving greater than 90% N removal in an SBR. During the first anoxic-fill stage, the influent
BOD is used as the donor for denitrification of the NO− 3 that remains with the settled biosolids from the previous cycle. The
second aerobic-reaction stage oxidizes the influent TKN to NO− 3 , while utilizing residual BOD aerobically. The third anoxic stage
denitrifies the remaining NO− 3 mainly through endogenous respiration of biosolids, releasing some NH+ 4
-N from the biosolids.
The fourth aerobic stage converts the NH+
4
-N produced from biosolids from the previous stage to NO−
3 , a part of which after
biosolids settling (Stage 5) is discharged to the effluent, the remainder being used in the first stage of the next cycle. The five-
stage SRB cycle has nearly the same functions in time as the five-stage Barnard process has in space. The important difference
is that the energy cost of the large fluid recycle for the Barnard process is avoided. Of course, two or more SBR reactors are
needed to handle the continuous influent flow.
Figure 13.5 A typical cycle (9 h) for achieving 90% N removal with a sequencing batch reactor.
Although the SBR cycle normally lasts for 8 to 12 h, SRT and HRT are similar to those of the Barnard process. The SRT is 15 d
or greater, while the HRT is around 24 h. SBR operation can be truncated to mimic only pre-denitrification, or the analog to the C
Systems in Figure 13.3. In this case, the anoxic fill, aerobic reaction, and settler-plus-draw stages are retained.
13.5.4. Side-Stream Anammox Treatment

13.5.4. Side-Stream Anammox Treatment
Although the anammox process for conversion of ammonium to N2 was not discovered until the 1990s, the full-scale
application of this unique process for side-stream denitrification was rapid, with at least 100 applications by 2014 (Lackner et
al., 2014). Side-stream treatment turned out to be an ideal location to apply the process’s first full-scale municipal wastewater
application, since the side-stream normally contains digester supernatant liquid with an ideal temperature for the slow-growing
anammox microorganisms, about 35°C, and a high ammonium-N concentration, typically 500 to 1500 mg/l. One of the major
challenges is to design a nitrogen-removal reactor for side-stream treatment that can hold the anammox organisms for the long
SRT required.
Table 13.8 indicates that, at 20°C, the minimum limiting SRT for anammox organisms is 10 d. At 35°C, it is much lower, say half
of that. Applying a safety factor of 4, a design SRT would then be about 20 d. What would be the best reactor design for such
side-stream anammox treatment? Several different designs have been used at full scale (Lackner et al., 2014). About 80% of the
systems surveyed used the DEMON® two-stage nitritification-anammox process, which includes continuous-feed SBRs, altering
between aeration with operational dissolved oxygen concentration of 0.3 to 0.5 mg/l for partial nitritification of ammonium to
nitrite, followed by anoxic anammox treatment for conversion of nitrite and ammonium to N2 . Here, aeration is activated when
the ammonium concentration exceeds an upper limit, and then is stopped when either the pH or ammonium concentration falls
below a lower limit. The low dissolved oxygen concentration generally used reduces the possible nitrite-formation rate, which, if
too high, might result in a too-high nitrite concentration, as well as unwanted conversion to nitrate (Wett, 2007). Nitrite, while a
necessary substrate for anammox organisms, can also be an inhibitor at higher concentrations. The control methods used with
DEMON plants are reported to maintain the nitrite-nitrogen concentration to a low and acceptable 5 mg/l (Wett, 2007).
Other side-stream reactor types and operational modes are intermittent vs. continuous feeding, suspended vs. attached
biomass, control of aeration, and one- vs. two-stage processes (Lackner et al., 2014). It is too early to determine which
approach will dominate in the long run, but it is clear that successful side-stream denitrification is effective.
The difficulties with side-stream treatment are that it provides only partial removal of ammonium-nitrogen for a full-scale plant
and the decrease in resources used is not as good as with traditional nitrification/denitrification. The main resource benefit
comes when the influent BODL/N ratio is too low for denitrification with recycle.
The next step with anammox is to find a successful approach for mainstream treatment, which, if it can be combined
successfully with efficient mainstream anaerobic treatment, could lead to a truly major transformation in wastewater treatment,
something sorely needed to better meet the resource and energy conservation requirements for future sustainability.
13.6. Nitrous Oxide Formation

Nitrous oxide, a greenhouse gas with a global warming potential that is 265 times that of CO2 (IPCC, 2013), can be formed
during nitrogen transformations. Massara et al. (2017) indicated that the influent N-fraction emitted as N 2 O from lab studies of
wastewater treatment varied from 0% to 95%, while full-scale treatment plants typically produce from 0% to 15%, the relative
amount being a function of several different factors. Surveying full-scale treatment plants, Ahn et al. (2010) found that up to
1.8% of influent TKN was emitted as N2 O, and the greatest emissions were consistently from aerobic zones.
As an example of the importance, consider the relative greenhouse gas contribution if just 5% of the influent N in the C1
recycle/denitrification system in Table 13.8 (part A) was discharged to the air as N2 O. The greenhouse gas impact would be
greater than that due to all of the plant’s other emitted CO2 equivalents, even if the plant’s energy needs came from a coal-fired
energy generator! In order to reduce the potential for N2 O formation, understanding the factors causing its production and how
to control them is important as we move toward sustainability.
Figure 13.6 illustrates processes by which N2 O production can occur during nitritification and denitrification; in both cases, N2 O
is a transformation intermediate.
Figure 13.6 The formation and decomposition of N2 O through nitritification and denitrification
pathways.
In the case of denitrification, N2 O is a natural intermediate following production of nitric oxide. Factors that slow down the
relative oxidation of ammonium to nitrite or reduction of nitrate or nitrite to N2 tend to increase the production of N2 O and to
reduce its decomposition. Possible factors here include a too-low ratio of electron donor to nitrate and an inhibitory dissolved
oxygen concentration. Cyclic feeding, such as in an SBR or a plug-flow reactor, may enhance N 2 O formation in this way
(Massara et al., 2017) and increase N2 O formation from less than 1% of influent N to a much higher fraction. The typical diurnal
variations in nitrogen and organic concentrations in wastewater streams can accentuate such variations.
In the case of ammonium oxidation to nitrite, N2 O can be formed via dimerization of NOH (nitroxyl) which decomposes to N2 O
(both at oxidation state +1). This can occur when NOH oxidation to NO and then to NO− 2 is blocked, say, by a low
concentration of dissolved oxygen as the terminal electron acceptor or possibly the accumulation of NO− 2 due to a lack of NOB
(Boiocchi et al., 2017; Kuypers et al., 2018; Massara et al., 2017). The imbalance may itself result from an insufficient dissolved
oxygen concentration for complete nitrification, a factor that can be accentuated by higher temperatures due to a greater
imbalance in the growth rates between AOB and NOB. N2 O production also might come from reduction of NO when its
oxidation to NO− 2 is blocked, such as by lack of O2 as an electron acceptor or accumulation of NO2 .
−
Although it is a moderately soluble gas, some of the N2 O formed escapes to the atmosphere, realizing its global warming
potential. The amount escaping depends on mixing and gas transfer conditions. The overall formation and fate of N 2 O
escaping to the atmosphere is a complex outcome of physical and microbiological factors.
Biofilms differ from suspended growth systems in the factors affecting N2 O formation and decomposition, because substrate
gradients and intermediate diffusion from one redox zone to another can be important. Sabba et al. (2017) indicated that N2 O
emissions from biofilms can be greater than from dispersed growth systems, depending upon bulk concentrations of oxygen,
COD, and nitrate, as well as biofilm thickness. For example, biofilms can be oxic in its surface layers, but anoxic in deeper
layers. Hence, N2 O can be formed within the biofilm through denitrification or even through nitrification closer to the surface at
a point where DO is sufficiently low, even when bulk conditions are aerobic. In contrast, a low bulk nitrate concentration can
lead to a low denitrification rate, which may lessen the accumulation of N2 O.
Another factor to consider is a unique approach that might make N2 O formation a benefit, instead of a liability. High levels of
N2 O formation can be obtained from nitrite reduction if the BOD concentration is maintained high compared with the
concentration of electron acceptors (Scherson et al., 2013). This leads to an entirely different nitrogen-removal process termed
CANDO, an acronym for “coupled aerobic-anoxic nitrous decomposition operation.” Here, wastewater’s ammonium is first
converted to nitrite; the nitrite is then fed to an anoxic reactor previously enriched with heterotrophic bacteria containing poly-3
hydroxybutyrate (PHB) through separate organic feeding. These microorganisms then can convert the nitrite to N 2 O. This N2 O
is then stripped as a gas and blended with the methane from anaerobic treatment of primary and secondary sludge and burned,
greatly enriching its energy content of the gas:
CH 4 + 4N2O → CO2 + 2H2O + 4N2, ΔH = −1219 kJ/mol
In summary, microorganism imbalance, variations in feeding frequency, and limiting concentrations of key substrates can lead
to increased N 2 O formation during nitrification and denitrification steps. While it might be possible in the future to benefit from
N2 O formation through the CANDO process, at the moment it is best to design systems to prevent N2 O formation. Keeping
sufficiently high the dissolved oxygen concentrations in nitrification reactors and organic donor concentrations in
denitrification are key steps to minimize net N2 O production. While the anammox process is sometimes claimed not to be
involved in N 2 O formation, nitritification must produce the required amount of nitrite, and this can lead to N2 O production
(Castro-Barros et al., 2015), perhaps even more so than with the conventional nitrification/denitrification process. Models able
to represent the various processes involved in N2 O formation are urgently needed.
13.7. References
Ahn, J. H.; S. Kim; H. Park; B. Rahm; K. Pagilla; and K. Chandran (2010). “N2 O emissions from activated sludge processes,
2008–2009: results of a national monitoring survey in the United States.” Environ. Sci. Technol. 45, pp. 4505–4511.
Barnard, J. L. (1975). “Biological nutrient removal without the addition of chemicals.” Water Res. 9, pp. 485–490.
Boiocchi, R.; K. V. Gernaey; and G. Sin (2017). “Understanding N2 O formation mechanisms through sensitivity analyses using
a plant-wide benchmark simulation model.” Chem Engg. J. 317, pp. 935–951.
Caranto, J. D. and K. M. Lancaster (2017). “Nitric oxide is an obligate bacterial nitrification intermediate produced by
hydroxylamine oxidoreductase.” Proc. Natl. Acad. Sci. U S A. 114, pp. 8217–8222.
Castro-Barros, C. M.; M. R. J. Daelman; K. E. Mampaey; M.C.M. van Loosdrecht, and E. I. P. Volcke (2015). “Effect of
aeration regime on N2 O emission from partial nitritation-anammox in a full-scale granular sludge reactor.” Water Res. 68, pp.
793–803.
Daigger, G. T. (2014). “Oxygen and carbon requirements for biological nitrogen removal processes accomplishing
nitrification, nitritation, and anammox.” Water Environ. Res. 86, pp. 204–209.
de Silva, D. G. V. and B. E. Rittmann (2000a). “Interpreting the response to loading changes in a mixed-culture completely
stirred tank reactor.” Water Environ. Res. 72, pp. 566–573.
de Silva, D. G. V. and B. E. Rittmann (2000b). “Nonsteady-state modeling of multispecies activated-sludge processes.”Water

Environ. Res. 72, 554–565.
Ekama, G. A. and M. C. Wentzel (2008). “Organic matter removal.” In Henze, M.; M. C. M. van Loosdrecht; G. A. Ekama; and
D. Brdjanovic, Eds. Biological Wastewater Treatment: Principles, Modelling and Design. London: IWA Publishing, pp. 87–138.
Furumai, H. and B. E. Rittmann (1994). “Evaluation of multiple-species biofilm and floc processes using a simplified
aggregate model.” Water Sci. Technol. 29, pp. 439–446.
Haug, R. T. and P. L. McCarty (1972). “Nitrification with submerged filters.”J. Water Pollut. Cont. Fed. 44, pp. 2086–2102.
Hellinga, C.; A. Schellen; J. W. Mulder; M. C. M. van Loosdrecht; and J. J. Heijnen (1998). “The SHARON process: an
innovative method for nitrogen removal from ammonium-rich waste water.” Water Sci. Technol. 37, pp. 35–142.
Herbold, C. W.; L. E. Lehtovirta-Morley; M.-Y. Jung; N. Jehmlich; B. Hausmann; P. Han; et al. (2017). “Ammonia-oxidising
archaea living at low pH: insights from comparative genomics.” Environ. Microbiol. 19, pp. 4939-4952.
IPCC (2013). Climate Change 2013: The Physical Science Basis. Cambridge, UK: Cambridge University Press.
Kim, J.; K. Kim; H. Ye; E. Lee; C. Shin; P. L. McCarty; et al. (2011). “Anaerobic fluidized bed membrane bioreactor for
wastewater treatment.” Environ. Sci. Technol. 45, pp. 576–581.
Kissel, J. C.; P. L. McCarty; and R. L. Street (1984). “Numerical simulation of mixed-culture biofilm.” J. Environ. Engg. ASCE
110, pp. 393–411.
Kits, K. D.; C. J. Sedlacek; E. V. Lebedeva; P. Han; A. Bulaev; P. Pjevac; et al. (2017). “Kinetic analysis of a complete nitrifier
reveals an oligotrophic lifestyle.” Nature 549, pp. 269–272.
Korner, H. and W. O. Zumft (1989). “Expression of denitrification enzymes in response to the dissolved oxygen level and
respiratory substrate in continuous culture of Pseudomonas stutzeri.” Appl. Environ. Microb. 55, pp. 1670–1676.
Konneke, M.; A. E. Bernhard; J. R. de la Torre; C. B. Walker; J. B. Waterbury; and D. A. Stahl (2005). “Isolation of an
autotrophic ammonia-oxidizing marine archaeon.” Nature 437, pp. 543–546.
Kuypers, M. M. M.; H. K. Marachant; and B. Kartal (2018). “The microbial nitrogen-cycling network.”Nat. Rev. Microbiol. 16,
pp. 263–276.
Lackner, S.; E. M. Gilbert; S. E. Vlaeminck; A. Joss; H. Horn; and M. C. M. van Loosdrecht (2014). “Full-scale partial
nitritation/anammox experiences—an application survey.” Water Res. 55, pp. 292–303.
Lee, H.-S.; Y. Tang; B. E. Rittmann; and H.-P. Zhao (2018). “Anaerobic oxidation of methane coupled to denitrification:
fundamentals, challenges, and potential.” Crit. Rev. Environ. Sci. Technol. 48, pp. 1067–1093.
Lee, K. C. and B. E. Rittmann (2002). “Applying a novel hollow-fiber biofilm reactor for autohydrogenotrophic denitrification
of drinking water.” Water Res. 36, pp. 2040–2051.
Lee, P. H.; W. Kwak; J. Bae; and P. L. McCarty (2013). “The effect of SRT on nitrate formation during autotrophic nitrogen
removal of anaerobically treated wastewater.” Water Sci. Technol. 68, pp. 1751–1756.
Lotti, T.; R. Kleerebezem; Z. Hu; B. Kartal; M. K. de Kreuk; C. V. E. T. Kip; et al. (2015). “Pilot-scale evaluation of anammox-
based mainstream nitrogen removal from municipal wastewater.” Environ. Technol. 36, pp. 1167–1177.
Martens-Habbena, W.; P. M. Berube; H. Urakawa; J. R. de la Torre; and D. A. Stahl (2009). “Ammonia oxidation kinetics
determine niche separation of nitrifying Archaea and Bacteria.” Nature 461, pp. 976–979.
Massara, T. M.; S. Malamis; A. Guisasola; J. Antonio Baeza; C. Noutsopoulos; and E. Katsou (2017). “A review on nitrous
oxide (N2 O) emissions during biological nutrient removal from municipal wastewater and sludge reject water.”Sci. Total
Environ. 596, pp. 106–123.
McCarty, P.L. (2018). “What is the best biological process for nitrogen removal: when and why?” Environ. Sci. Technol. 52,
pp. 3835–3841.
McCarty, P. L.; J. Bae; and J. Kim (2011). “Domestic wastewater treatment as a net energy producer—can this be achieved?”
Environ. Sci. Technol. 45, pp. 7100–7106.
McCarty, P. L.; L. Beck; and P. St. Amant (1969). “Biological denitrification of wastewaters by addition of organic materials.”
Extension Series 135. West Lafayette, IN: Purdue University, pp. 1271–1285.
Mobarry, B. K.; M. Wagner; V. Urbain; B. E. Rittmann; and D. A. Stahl (1996). “Phylogenetic probes for analyzing abundance
and spatial organization of nitrifying bacteria.” Appl. Environ. Microbiol. 62, pp. 2156–2162.
Owen, W. F. (1982). Energy in Wastewater Treatment. Englewood Cliffs, NJ: Prentice-Hall.
Rittmann, B. E. (2018). “Biofilms, active substrata, and me.” Water Res. 132, 135–145.
Rittmann, B. E. and J. A. Manem (1992). “Development and experimental evaluation of a steady-state, multispecies biofilm
model.” Biotechnol. Bioengg. 39, pp. 914–922.
Rittmann, B. E.; J. M. Regan; and D.A. Stahl (1994). “Nitrification as a source of soluble organic substrate in biological
treatment.” Water Sci. Technol. 30(6), pp. 1–8.
Rittmann, B. E. and W. E. Langeland (1985). “Simultaneous denitrification with nitrification in single-channel oxidation
ditches.” J. Water Pollution Control Fedn. 57, pp. 300–308.
Rittmann, B. E. and V. L. Snoeynik (1984). “Achieving biologically stable drinking water.”J. Am. Water Works Assoc. 76(10),
pp. 106–114.
Sabba, F.; C. Picioreanu; J. P. Boltz; and R. Nerenberg (2017). “Predicting N2 O emissions from nitrifying and denitrifying
biofilms: a modeling study.” Water Sci. Technol. 75, pp. 530–538.
Scherson, Y. D. and C. S. Criddle (2014). “Recovery of freshwater from wastewater: upgrading process configurations to
maximize energy recovery and minimize residuals.” Environ. Sci. Technol. 48, pp. 8420–8432.
Scherson, Y. D.; G. F. Wells; S.-G. Woo; J. Lee; J. Park; B. J. Cantwell; et al. (2013). “Nitrogen removal with energy recovery
through N2 O decomposition.” Energy Environ. Sci. 6, pp. 241–248.
Shin, C.; J. Bae; and P. L. McCarty (2012). “Lower operational limits to volatile fatty acid degradation with dilute
wastewaters in an anaerobic fluidized bed reactor.” Biores. Technol. 109, pp. 13-20.
Stahl, D. A. and J. R. del la Torre (2012). “Physiology and diversity of ammonia-oxidizing Archaea.”Ann. Rev. Microb. 66, pp.
83–101.
Tang, Y.; C. Zhou; S. Van Ginkel; A. Ontiveros; J. Shin; and B. E. Rittmann (2012). “Hydrogen-permeabilities of the fibers used
in a H2 -based membrane biofilm reactor.” J. Membrane Sci. 407–408, pp. 176–183.
Tiedje, J. M. (1988). “Ecology of denitrification and dissimilarity nitrate reduction to ammonium.” In A. J. B. Zehnder, Ed.
Biology of Anaerobic Microorganisms. New York: John Wiley, pp. 179–244.
van Kessel, M. A. H. J.; D. R. Speth; M. Albertsen; P. H. Nielsen; H. J. M. Op den Camp; B. Kartal; et al. (2015). “Complete
nitrification by a single microorganism.” Nature 528, pp. 555–559.
Weissbach, M.; P. Thiel; J. E. Drewes; and K. Koch (2018). “Nitrogen removal and intentional nitrous oxide production from
reject water in a coupled nitritation/nitrous denitritation system under real feed-stream conditions.” Biores. Technol. 255, pp.
58–66.
Wett, B. (2007). “Development and implementation of a robust deammonification process.” Water Sci. Technol. 56(7), pp.
81–88.
Winkler, M. K. H.; J. Yang; R. Kleerebezem; E. Plaza; J. Trela; B. Hultman; et al. (2012). “Nitrate reduction by organotrophic
anammox bacteria in a nitritation/anammox granular sludge and a moving bed biofilm reactor.” Biores. Technol. 114, pp.
217–223.
Ziv-El, M. C. and B. E. Rittmann (2009a). “Water-quality assessment after treatment in a membrane biofilm reactor.J. Am.
Water Works Assoc. 101(12), pp. 77–83.
Ziv-El, M. C. and B. E. Rittmann (2009b). “Systematic evaluation of nitrate and perchlorate bioreduction kinetics in
groundwater using a hydrogen-based membrane biofilm reactor.” Water Res. 43, pp. 173–181.
13.8. Problems
13.8. Problems
13.1 A nitrification process can easily be operated in a fixed-film mode. Typical kinetic parameters are as follows:
K = 0.5 mg NH+
4
-N/l Df = 1.2 cm2/d Υ = 0.26 mg VSa/mg NH+
4
-N L = 65 µm
q̂ = 2.0 mg NH+
4
-N/mg VSa-d D = 1.5 cm2/d b' = 0.1/d Xf = 5 mg VSa/cm3
You are to determine the response of a complete-mix biofilm reactor to changes in detention time. The reactor has a
specific surface area, a, of 1.0 cm−1. The influent feed strength (S 0 ) is 30 mg NH+
4
-N/l. You can assume that NH+
4
-N
completely limits the kinetics.
The steps for the solution are as follows:
a. Calculate S bmin, S *bmin, and JR.
b. Generate a curve of J/JR versus S/S min.
c. Set up the mass balance for substrate in the reactor.
d. For each detention time (1, 2, 4, 8, and 24 h), solve for S. You may wish to use an iterative technique. Guess S, solve for J
(from curve from part b), calculate S from the mass balance, check for agreement of S, and repeat until convergence.
Ignore suspended reactions.
e. Plot S versus θ. Interpret the curve.
f. Plot S 0 /S versus θ and see if an apparent first-order reaction seems like a good description of the process. If so, what is
the apparent first-order coefficient? If not, why is the reaction not first-order?
13.2 You are to design a fixed-film biological process to oxidize ammonia nitrogen. The wastewater contains 50 mg NH+
4
-N/l in a flow of 1000 m3 /d. Available are modules of complete-mix biofilm reactors; each module has 5000 m2 of surface
area for biofilm colonization. How many modules, operated in series, are needed to achieve an effluent concentration of 1
mg NH+ 4
-N/l or less? You may use the following parameters and may assume that NH+ 4
-N is the rate limiting substrate.
K = 1.0 NH+
4
-N/l Df = 1.3 cm2/d Υ = 0.33 mg VSa/mg NH+
4 -N bdet = 0.1/d
q̂ = 2.3 mg NH+
4
-N/mg VSa-d D = 1.5 cm2/d b = 0.11/d Xf = 40 mg VSa/cm3
L = a value so small that Ss approaches S.
What do you expect the actual effluent concentration to be?
13.3 A set of kinetic coefficients for nitrification is as follows:
Species q̂ kg NH +
4
-N/kg VSSa-d K mg NH +
4
-N/l b d−1 Y kg VSS a/kg NH +
4
-N fd
Ammonium oxidizers 1.4 0.7 0.05 0.5 0.8
Nitrite oxidizers 9 2.2 0.05 0.1 0.8
If nitrification is to work reliably, the apparent safety factor (SF) for activated sludge will be at least 20 for the lowest
growing species. If SF = 20, what will be the sludge production rate of active, inactive, and total VSS (kg/d) if the NH+ 4
-N
removal is 367 kg N/d? Note that you must consider each species separately. You also may ignore SMP and heterotrophs.
13.4 Use the kinetic parameters in Table 13.3 to calculate N1 and θ for an activated sludge nitrification process (complete-
mix reactor with settler and recycle) treating 500 mg/l of NH+
4
-N with θx = 15 d and MLVSS = 2500 mg/l. Assume no
influent solids or BOD. Calculate the N1 and N2 values for 5, 10, 15, and 20°C. Hint: First calculate S values for NH+
4
-N and
NO−
2 , then calculate Xv for both species of nitrifiers. Ignore SMP. Do not ignore N used for synthesis.
13.5 A treatment plant was designed to carry out nitrification, as well as BOD removal, in a single-stage system. In order to
assure that nitrification occurred, the designers used a low loading rate of 0.2 kg BOD5 /m3 -d. Nonetheless, nitrification
performance has been poor. You are the high-priced consultant and have obtained the following operating data: water
temperature 15°C, aeration basin D.O. = 2.5 mg/l, aeration basin regime essentially completely mixed, MLVSS = 1200 mg/l,
Q = 104 m3 /d, reactor volume = 1.25(104 ) m3 .
Influent quality: BOD5 = 250 mg/l, total Kjeldahl N = 23 mg/l
Effluent quality: BOD5 = 25 mg/l, VSS = 25 mg/l, NH+

4
-N = 20 mg N/l
Recycle sludge quality: recycle ratio = 0.32, volatile solids content = 5(103 ) mg/l
Waste sludge flow rate: 9(102 ) m3 /d
What is the problem and how can it be solved?
13.6 Your client wants to treat his ammonium-containing wastewater by a trickling filter nitrification process to eliminate
the nitrogenous oxygen demand. The water is at 20°C, and you have estimated the limiting kinetic coefficients to be
q̂ = 2.6 mg NH+
4
-N/mg VSSa-d K = 3.6 mg NH+
4
-N/l
Y = 0.26 VSSa/mg NH+
4
-N
Yobs = 0.2 mg VSSa/mg NH+
4
-N b = 0.05/d
The process configuration that your client wants includes Q = 10,000 m3 /d, N10 = 50 mg NH+
4
-N/l, a = 50 m2 /m3 , depth of
the filter = 2 m, and volume of filter medium = 500 m3 .
Does it seem feasible that the client’s design can achieve the desired
95% removal with these conditions? You will need to consider N10min and O2 flux first. Compare necessary N1 and JO2 with
N10min and the practical JO2 (about 2.85 mg O2/cm 2-d). Are they reasonable? If yes, is the volume reasonable? You may
ignore heterotrophs here.
13.7 RBCs are designed for nitrification with two loading criteria. First, the BOD load must be low enough to allow nitrifiers
to compete. Second, the NH+
4
-N flux must be sufficient to give adequate N oxidation. Estimate the required surface area if
the BOD load is 10 kg BOD5 /l000 m2 -d for the carbonaceous stages, but less than 4 kg BOD5 /l000 m2 -d for the whole
process, and the NH+ 4
-N load is 0.4 kg N/l000 m2 -d in the nitrifying stages. The flow is 0.1 m3 /s, influent BOD5 = 250 mg/l,
and influent TKN = 50 mg N/l.
13.8 An existing wastewater treatment plant has a reactor volume of 2000 m3 . The clarifier has a volume of 750 m3 and a
surface area of 300 m2 . Reactor aeration is supplied by 100 kW brushes that have a field OTE of 1.2 kg O2 /kWh. The
elevation is sea level. The desire is to operate the plant with an extended aeration θx of 30 d. You may assume the following
kinetic parameters:
Process q̂ K b Y fd
BOD oxidation 20 mg BOD L/l 0.2/d 0.8

16 g BODL 0.6 g VSSa
g VSSa ⋅ d g BODL
Nitrification 0.05/d 0.8

1.7 g NH +
4
-N 0.5 mg NH +
4
-N/l 0.6 g VSSa
g VSSa ⋅ d g NH +4
-N
(total oxidation)
Let T = 15°C. Use a 1.1 g TSS/g VSS ratio in design. Let the DO concentration be 2 mg/l andXv0 = 0. For simplicity, you can
assume 100% oxidation of NH+ 4
-N and BOD L. Estimate the maximum practical BODL loading (kg BODL/d) and the maximum
flow rate (m 3 /d), assuming the maximum BODL load is obtained when the g BODL/g TKN in the influent is 10.
13.9 Operation of the nitrification plant as a single-stage or two-stage system can affect the total O2 consumption and
total sludge production. In this problem you will compare a single-stage system having an SRT of 15 d with a two-stage
system that has SRTs of 3 and 15 d, respectively.
Kinetic parameters are as follows:
Process q̂ K b Y fd
BOD oxidation 20 mg BOD L/l 0.2/d 0.8

g VSSa ⋅ d g BODL
Nitrification 0.05/d 0.8

1.7 g NH +
4
-N 0.5 mg NH +
4
-N/l 0.4 g VSSa
-N
The flow rate is 1000 m3 /d, and influent concentrations of BODL and TKN are 400 and 40 mg/l, respectively. Assume no
input solids, and ignore products.
You are to take a stepwise approach:
a. For the first stage of the two-stage system:
1. Calculate effluent BODL concentration.
2. Calculate total kg BODL removed per day.
3. Calculate the VSS production rate per day.
4. Calculate the kg N/d incorporated into cells.
5. Calculate the kg O2 /d consumed for heterotrophic reactions.
b. For the second stage of the two-stage system:
1. Calculate the effluent TKN concentration.
2. Calculate the kg/d of TKN oxidized.
3. Calculate the production rate of VSS from nitrification.
4. Calculate the kg O2 /d consumed in nitrification.
c. For the single-stage system:
1. Calculate the effluent BODL concentration.
2. Calculate the kg BODL/d removed.
3. Calculate the VSS production rate from heterotrophic reactions.
4. Calculate the rate of incorporation of N into heterotrophic cells.
5. Calculate the O2 consumption rate for heterotrophic reactions.
6. The effluent TKN concentration is the same as before. Calculate the total TKN removed by nitrification.
7. Calculate the VSS produced per day via nitrification.
8. Calculate the kg O2 /d consumed via nitrification.
d. Prepare a table with separate columns for the single-stage and two-stage systems. Each column should have entries of
VSS production rate and O2 consumed for BOD oxidation, nitrification, and total.
13.10 Estimate the required volume of a Bicarbone (also called BAF) biofilm process to treat wastewater that has a flow
rate of 500 m3 /d, and influent NH+
4
-N concentration of 100 mg N/l, and a temperature of 25°C. You can use the ammonium
oxidizers’ kinetic parameters from Table 13.3. Assume bdet = 0.1 d−1, that you wish to achieve an effluent concentration of 1
mg NH+ 4
-N/l, a porosity of 0.4, L = 50 µm, and Xf = 40 mg/cm3 . You should assume that the reactor is moderately well
mixed by aeration, such that it has two segments in series.
13.11 A reasonable estimate for θx is 1/bdet for a biofilm process. A plug-flow, fixed-bed nitrification process is to have an
effluent concentration of 0.2 mg NH+
4
-N/l when it is operated in the low load region at a temperature of 15°C. At
approximately what SRT is the biofilm? Does this value make sense? Use the 15°C kinetic parameters for ammonium
oxidizers in Table 13.3 to make your analysis.
13.12 You are attending a meeting with your client who operates a two-stage trickling filter system. The following sketch
is shown to you:
The two trickling filters are of the biological-tower type. They each have a depth of 8 m, a diameter of 16 m, and contain
plastic media having specific surface area of 200 m−1. Each clarifier is 3 m deep and has a diameter of 32 m. In addition,
you are given the following information: influent flow rate Q = 10,000 m3 /d, influent BOD5 = 600 mg/l, and influent TKN = 100
mg/l. You have only a few minutes to give your expert opinion concerning whether or not this system ought to be able to
successfully achieve BOD oxidation and nitrification. Make a rapid (but relevant and expert) analysis of the system. Explain
why the system should or should not achieve the goal. Use the first-stage and overall BOD and TKN loads.
13.13 Bioaugmentation is a relatively recent concept for biological processes. The idea is to add a biological material,
such as special bacteria or enzymes, to enhance the performance of the process. One application is nitrification. Here,
nitrifying bacteria are added to the influent of a reactor to “bioaugment” nitrification. You are to analyze how well
bioaugmentation will work to upgrade an aerated lagoon. The lagoon’s current parameters are as follows:
Flow rate: 1000 m3 /d
Influent BOD L: 20 mg/l
Volume: 3000 m3
Temperature: 10°C
Influent NH+
4
-N: 200 mg/l
D.O.: 4 mg/l
Effluent NH+
4
-N: variable, but mostly near 200 mg/l
The bioaugmentation materials consist of freeze-dried Nitrosomonas. The cost is $l00/kg VSS, and you can assume that the
VSS are 100% active. How much money will it cost to bioaugment with the Nitrosomonas materials if you are to have a
stable effluent concentration of 0.1 mg N/l of NH+
4
-N?
You can use the kinetic parameters listed inTable 13.3 for 10°C. You may ignore any heterotroph growth and SMP. Assume
that the oxygen supply is adequate.
13.14 You wish to design a single-stage activated sludge process to nitrify. The influent has these characteristics: BODL =
250 mg/l, TKN = 100 mg/l, NO−3 = 40 mg N/l, inert VSS i = 35 mg/l, and alkalinity = 1000 mg/l as CaCO3 . Determine the
O2 demand (in mg O2 /1 of wastewater) if the following parameters are valid:
Process qx q̂ K b Y fd
Heterotrophs 15 d 10 mg BOD L/l 0.15/d 0.8

g VSSa ⋅ d g BODL
Nitrifiers 15 d 0.15/d 0.8

2.0 g NH +
4
-N 1.5 mg NH +
4
− N/l 0.41 g VSSa
-N
13.15 A biological tower has a height of 10 m, a square cross section of 5 m × 5 m, and a modular media witha = 200 m−1.
It receives influent having a flow rate of 350 m 3 /d, 600 mg BODL/l, and 50 mg TKN/l. Effluent recycle is at 350 m3 /d.
Evaluate whether or not this process ought to be successful at nitrification and BOD removal.
13.16 You have a wastewater with the following characteristics: Q = 2000 m3 /d, BODL = 400 mg/l, TKN = 50 mg/l, and
NO−
3 -N + NO− 2 -N = 0 mg/l. You decide to use a multi-stage RBC to achieve excellent BOD and NH4 removals. You
+
design the first stage just for BOD oxidation. Its surface loading is the “traditional” first-stage one: 45 kg BOD5 /1000 m2 -d,
which converts to 66 kg BOD L/1000 m2 -d. You then follow that first stage with three more stages of exactly the same “size.”
You have estimated that the BODL concentration will decline along the RBC stages according to the following:
Location Influent Stage 1 Effluent Stage 2 Effluent Stage 3 Effluent Stage 4 Effluent
BOD L (mg/l) 400 135 50 30 20
Will this RBC system achieve excellent removal of NH+

4
-N? If yes, in which stages will it occur? If not, why isNH+
4
-N
removal a failure?
13.17 RBCs have been used for denitrification, as well as for aerobic treatment. Estimate the surface area needed to
remove essentially all 30 mg NO− 3 -N/l from a wastewater by a complete-mix RBC having a temperature of 20°C, methanol
is the carbon source and limiting substrate, the effluent methanol concentration is 1 mg/l, and Q = 1000 l/d. You may
assume that the average SRT is 33 d to calculate the stoichiometry, which you may assume is constant. Also, you may
assume steady-state operation, and the following parameters:
K = 9.1 mg CH3OH/l Df = 1.04 cm2/d bdet = 0.03 d− 1

b = 0.05 d− 1 L = 60 µm q̂ = 6.9 mg CH3OH/mg VSa-d
D = 1.3 cm2/d Xf = 20 mg VSa/cm3 Υ = 0.27 mg VSa/mg CH3OH
Use the steady-state-biofilm model directly.
13.18 A set of kinetic coefficients for denitrification in a CSTR is as follows:
Substrate q̂ K b Y
Methanol 15 mg methanol/l 0.05/d

8.3 g methanol 0.27 g VSSa
g VSSa ⋅ d g methanol
Nitrate 0.05/d
2.8 g N 0.1 mg N/l 0.81 g VSSa
g VSSa ⋅ d gN
a. Calculate the steady-state concentrations of methanol and nitrate-N for θx = 1, 2, 3, 4, and 5 d, assuming the respective
substrate is limiting. In other words, there are two answers for each θx.
b. If you have an initial concentration of 20 mg/l of NO− 3 -N, what is the maximum concentration of methanol that may be
present in the feed to preclude nitrate limitation for each θx.
13.19 An industrial wastewater is characterized as follows: BODL = 5000 mg/l, TKN = 150 mg/l,
NO−
3 -N + NO− 2 -N = 0 mg/l, SS = 250 mg/l, pH = 8.4, alkalinity = 2000 mg/l, and TDS = 4000 mg/l. Evaluate the
technical and economic suitability of using a first-stage denitrification scheme with second-stage aerobic treatment for this
wastewater. In other words, is it possible to do denitrification first? Compared to just aerobic oxidation, what are the effects
of first-stage denitrification on O 2 usage, sludge production, and pH?
13.20 Denitrification is being used for BOD and N removal in a first-stage anoxic reactor. The flow then goes to the
nitrification reactor; a clarifier follows the nitrification reactor, and sludge recycle is to the head of the denitrification reactor.
The θx value is 15 d. The coefficients are as presented in Table 13.9 for the denitrification and aerobic organic oxidation in
System C, and ammonia oxidation in the second-stage reactor. Your ultimate goals are to determine the sludge production
rate and the oxygen-transfer rate. The influent BODL is 300 mg/l. The influent TKN is 25 mg/l. NO− 3 -N is negligible in the
3 3
feed. The influent flow rate is 770 m /d. The mixed-liquor recycle rate is 7700 m /d, and the sludge recycle rate is 385
m3 /d. To get your final result, take the following steps:
a. Calculate the amount of BODL removed by denitrification. The amount of BODL removed by denitrification is proportional
to the NO− −
3 -N removed ultimately as N2 . Assume that all NH4 -N is nitrified to NO3 .
+
b. Calculate the amount of BODL removed in the aerobic nitrification tank. The BODL removal aerobically is that BODL
remaining in the influent to the aerobic tank minus the soluble effluent BODL, which can be assumed to be 2.5 mg/l.
c. Calculate the cells produced (kg VSS/d) via denitrification.
d. Calculate the cells produced via aerobic BOD L oxidation.
e. Calculate the cells produced during nitrification.
f. Calculate the total cell production rate.
g. Calculate the O2 required (kg/d) for nitrification and BODL oxidation, and then total the results.
13.21 List six advantages and/or disadvantages of using denitrification before nitrification (with recycle ofNO−
3 ), instead
of the three-stage approach for sequential BOD oxidation, nitrification, and denitrification.
13.22 An industrial waste contains nitric acid (HNO3 ) at 7 mM. An aspiring young environmental engineer, eager to
impress her new employer, suggests the employer uses waste syrup as the carbon source. Syrup has a BODL of 100,000
mg/l and is basically pure carbohydrate (C6 H12O6 ). The engineer says to operate the system with θx = 8 d. If fd = 0.8,
fs0 = 0.6, b = 0.08 d−1, q̂ = 18 g BODL/g VSS a-d, and K = 20 mg BODL/l, how much syrup (volume) must be added to
remove essentially all the nitric acid? What is the absolute minimum amount of alkalinity needed to neutralize the acid in the
treated wastewater?
13.23 A wastewater has the following characteristics: BODL = 2000 mg/l, TKN = 59 mg/l, NO− 3 -N = 1 mg/l,
3− + + 2+ 3−
PO4 -P = 10 mg/l, K = 78 mg/l, Na =170 mg/l, Ca = 200 mg/l, SO4 = 500 mg/l, Cl = 300 mg/l, Mg2+ = 49 mg/l,
−
and alkalinity as CaCO 3 = 200 mg/l. Select a biological process or series of biological processes to reduce the BODL to 20
mg/l and the total N to about 1 mg/l. What should θx or θ (as appropriate) be? Must there be any chemical additions? If so,
what and how much? What is the O2 utilization? How much sludge is produced?
13.24 You wish to design a denitrification process that will remove the nitrate-N in a wastewater to less than 1 mg N/l.
Your source of BOD is spent-grains liquor from a brewery. Not only must you achieve virtually 100% N removal, but you must
also achieve an effluent BOD L of 5 mg/l or less from the influent BOD. The influent has 50 mgNO−3 -N/l; no nitrite, ammonia,
3
or BOD; and a flow rate of 100 m /d. You can use the following kinetic parameters for T = 20°C.
Substrate q̂ K b Y fs0 D Df Xf
BOD L 0.36
10.4 g BODL 50 mg BODL 0.05 0.18 g VSa 0.6 cm 2 0.48 cm 2 20 mg VSa
g VSa ⋅ d l d g BODL d d cm 3
Nitrate 0.36
2.3 g N 0.1 mg NO−
3 −N
0.05 0.81 g VSa 1.0 cm 2 0.8 cm 2 20 mg VSa
g VSa ⋅ d l d g NO− d d cm 3
3 −N
You chose to use a fluidized-bed biofilm reactor having a once-through liquid regime, 2 mm sand for media, and fluidization
to 1.5 times the unfluidized bed height. Your overall goal is to determine a feasible design size (i.e., empty-bed contact
time), if it is possible to achieve your treatment goals.
a. Determine ∈ and ó for fluidized conditions, assuming the sand has a specific gravity of 2.65, a nonfluidized porosity of
0.35, and can be considered as a sphere.
b. Estimate S min for NO−

3 -N and BOD L. Choose the process-limiting substrate.
c. Select an appropriate loading criterion for the process-limiting substrate. If the design is not feasible, state the reasons
for infeasibility at this point and then select an appropriate load to come as close as possible to the desired effluent
quality. (You may assume L = 50 µm.)
d. Estimate the biofilm thickness and check your estimate of S min. If you must adjust S min, describe qualitatively how the
change would affect your criterion or performance, but do not rework the problem. Continue on to part (e) using the
same design value.
e. Calculate the required influent BODL concentration.
f. Calculate the fluidized-bed empty-bed detention time for your design.
13.25 Consider a fluidized-bed denitrification reactor that has an empty-bed detention time of 10 min. The original medium
is 0.2-mm sand particles having a packed porosity of 0.25. In operation, the bed is expanded to 1.6 times its packed height.
The feed strength is 100 mg/l of methanol and 25 mg NO− 3 -N/l.
Calculate the following:
a. Fluidized-bed porosity.
b. Specific surface area of fluidized-bed (assume 2 mm spheres).
c. Shear stress in reactor (based on fluidization of sand at a density of 2.65 g/ml).
d. Approximate biofilm loss coefficient due to shearing and decay, b′.
e. S bmin for methanol. Use b′ and kinetic parameters listed in Table 13.8.
f. If methanol is limiting and its effluent concentration approaches S min, what is the “average” methanol flux in the reactor?
What is the “average” biofilm thickness if Xf = 40 mg/cm3 ? What is the effluent NO− 3 -N concentration?
13.26 The operator of an activated sludge treatment plant was puzzled when an engineer from the state EPA indicated
that the activated-sludge plant was improperly designed and could not possibly provide adequate treatment of the BODL
load. The EPA engineer stated that the operator must install twice as much aeration capacity to make the plant work. The
operator was puzzled because effluent standards have been met for several years. The effluent standards are as follows:
20 mg BOD5 /l, 20 mg SS/l, and 1.5 mg NH+ 4
-N/l. In addition, the engineer noted that D.O. was present in the mixed liquor.
The activated sludge plant has the following specifications: aeration tank volume = 2500 m3 , settler area =330 m2 , aerator
capacity = 65 kW. The influent characteristics to the activated-sludge unit are as follows: flow rate = 10,000 m3 /d, BOD5 =
200 mg/l, SS = 75 mg/l, TKN =15 mg/l, NO− 3 -N = 40 mg/l, and P = 12 mg/l. The mixed liquor has 4000 mg SS/l. Waste
sludge has 9000 mg SS/l and is wasted at rate of 75 m3 /d. An aeration test gave an aerator field oxygen transfer efficiency
(FOTE) of 1 kg O2 /kWh under actual conditions.
You are called in as the high-priced consultant to help the operator defend himself. You have only a few minutes to analyze
the situation and present a lucid and correct explanation of why the plant works properly when the EPA engineer says that it
is under-designed with respect to oxygen transfer capacity. You need not make a highly detailed analysis, but you must be
able to quantify your explanation. Therefore, demonstrate if and why the EPA engineer has made an incorrect analysis.
13.27 Denitrification is sometimes used to remove nitrate from drinking water. Provide a preliminary design for a fixed-film
process to remove 10 mg NO 3 −-N/l to less than 0.1 mg/1. The preliminary design involves estimating the total reactor
volume and the amount of added acetate needed as the electron donor. The known information is:
Substrate q̂ K b bdet Y fd D Df Xf
Acetate 10 g AC− 15 mg AC− 0.1 0.05 0.18 g VSa 0.8 1.2 cm2 1.0 cm2 40 mg VSa
g VSa ⋅ d 1 d d g AC − d d cm 3
Nitrate 2.3 g NO−

3 -N 0.1 mg NO−
3 -N
0.1 0.05 0.81 g VSa 0.8 1.2 cm2 1.2 cm2 40 mg VSa
g VSa ⋅ d d d g NO − d d cm 3
l 3 -N
3 −1
You a1so know that Q = 1,000 m3 /d, a = 300 m−1, L = 90 μm. You may assume that the added acetate is the rate-limiting
substrate, as long as it is not fully depleted. Using the steady-state-biofilm model, you will need to estimate S min for each
substrate and the amount of acetate needed to remove the 10 mg NO3 −–N/l. Then, determine the needed J and volume to
keep residual acetate to as low a level as practical.
13.28 You need to design a biofilm process to be used as a first-stage denitrification scheme that uses only biofilm
processes. The wastewater to be treated has the following characteristics: Q = 4,000 m3 /d, BODL = 300 mg/l, TKN = 65 mg
N/l, NO3 −-N + NO2 −-N = 0 mg N/l. Your design calls for 85 percent removal of N, and you have correctly figured out that the
recycle flow rate from a second-stage (aerobic) biofilm reactor is 22,700 m3 /d. Also, you are able to assume that N uptake
in cell synthesis is not important. Furthermore, pilot studies with denitrifying biofilm reactor of a similar type gave values of
bdet = 0.04 d−1, Xf = 40 mg/cm3 , and L = 40 μm.
Using the steady-state-biofilm model, your goal is to determine the total reactor volume of a completely mixed biofilm
reactor when the specific surface area is 200 m−1. Some potentially important kinetic parameters for the rate-limiting
substrate for denitrification are:
q̂ = 12 mg BODL/mg VSSa-d K = 20 mg BODL/1 b = 0.05 d− 1

Y = 0.27 mg VSSa/mg BODL D = 0.6 cm2/d Df = 0.48 cm2/d
13.29 Your firm needs to design a process to treat a wastewater with a somewhat unusual composition. The key features
are as follows: BODL = 250 mg/l, TKN = 50 mg/l, NO− 3 -N = 50 mg/l, and SS = 0 mg/l. Your boss wishes to use an oxidation
ditch to treat this wastewater by the technique of simultaneous nitrification with denitrification in a single reactor. The boss
assigns to you the task of assessing the feasibility of such an approach. In particular, you are to assess whether or not you
can accomplish nearly complete BOD and N removals with this wastewater in such a process. You brilliantly realize that the
one reactor system can be represented as a first-stage denitrification system with a very high mixed-liquor recycle ratio, say
1000. Could such a process be successful?
13.30 One potential benefit of biofilm processes is that they can achieve a very high volumetric loading, which makes the
process compact. To achieve a high volumetric loading, you need a high specific surface area, such as with a fluidized-bed
biofilm reactor. Some say that denitrification is the ideal reaction for a compact fluidized-bed process. Give two aspects of
denitrification that make it a good choice for a compact biofilm process.
13.31 You are going to treat a wastewater having the following characteristics: BODL = 500 mg/l, TKN = 50 mg/l, inert SS =
50 mg/l, and NO− 3 -N = 0 mg/1. You have five options for treatment, as shown schematically below. You are to indicate for
each comparison how biosolids production (in kg VSS/d) and the oxygen required (in kg O2 /d) change by making the
changes indicated in the comparisons. For example, how do the sludge production and oxygen required change when you
go from process (a) to process (b), which is the first comparison? Your choices in all cases are as follows: increases,
decreases, or no change. Comparisons to make are 1. a to b, 2. a to c, 3. b to d, and 4. b to e.
13.32 First-stage denitrification ought to be achievable using biofilm reactors, instead of suspended-growth reactors.
Draw a schematic diagram of such a system. Label the five most important parts of the system. Describe what each part
does and why it is necessary.
13.33 You are the lead engineer for the design of a two-stage aerobic-denitrification system with the first-stage used for
denitrification. The somewhat unusual wastewater characteristics are as follows: BODL = 2000 mg/l, TKN = 300 mg/l, and
NO− 3 -N = 150 mg/l. Your first design decision is to set the SRT at 20 d. At this SRT, you can assume that the anoxic reactor
will drive the NH+ 4
-N and BOD L concentrations so close to zero that you can take them as zero.
a. You need first to assure yourself that the influent BOD is sufficient to drive complete denitrification. Is it?
b. What mixed-liquor recycle ratio is required to achieve a total N concentration in the influent of 30 mg/l?
c. How much of the BODL (expressed in milligrams per liter in the influent flow) will be removed by denitrification and by
aerobic oxidation?
13.34 You wish to consider the addition of methanol to groundwater for anaerobic biological removal of nitrate by
denitrification. If the nitrate-nitrogen concentration is 84 mg/l, what minimum concentration of methanol should be added to
achieve complete nitrate reduction to nitrogen gas? Assume no ammonia is present and that fs for the reaction is 0.30.
13.35 You wish to design a plug-flow biofilm reactor without recycle for denitrification to N2 of a wastewater containing 65
mg NO−3 -N/l and no ammonia, and you have selected acetate to add as the electron donor for the reaction.
a. What concentration of acetate should be added to the wastewater to achieve complete removal of nitrate by
denitrification?
b. Given the above concentration of acetate added to the wastewater prior to treatment, is the reaction at the entrance to
the reactor rate-limited by acetate, nitrate, or neither? Show appropriate calculations to support your conclusion. Assume
that nitrate is used for cell synthesis, and that the following characteristics apply for reaction kinetics and biofilm
characteristics:
Constituent Acetate Nitrate Biofilm
K 10 mg/l 1 mg/l
q̂ 15 mg/mg VS a-d
Dw 0.9 cm2/d 0.7 cm2/d
Df 0.8 Dw 0.8 Dw
L 0.150 cm
fs 0.55
Xf 12 mg/cm3
b 0.15/d
Y 0.4 g VS a/g acetate
13.36 A biofilm reactor is used for denitrification of a water supply, and methanol is being considered as an electron donor
for the reaction. For the following conditions, estimate the flux rate for nitrate into the biofilm (mg/cm2 -d). Assume NH3 -N is
available for cell synthesis and use the steady-state-biofilm model:
Constituent Methanol Nitrate-N Biofilm
S0 1 mg/l 3 mg/l
K 15 mg/l 3 mg/l
q̂ 10 mg/mg VS a-d 3.2 mg.mg VS a-d
D 1.3 cm2/d 0.7 cm2/d
Df 0.8 D 0.8 D
L 0.090 cm
Xf 15 mg VS a/cm3
Y 0.42 mg VS a/mg
b' 0.1/d
13.37 You are designing a CSTR with recycle for anoxic denitrification of an industrial wastewater containing nitrate, and
are considering the addition of thiosulfate (S2 O3−−), a waste chemical at the plant, to be used as the electron donor for the
biological reaction. As such the thiosulfate would be oxidized to sulfate. You are conducting a sensitivity analysis for the
design to determine the influence of different variables on reactor size, nitrate concentration in the reactor effluent, and
thiosulfate addition required. Since nitrate is the contaminant of interest, you desire to keep it, rather than thiosulfate, as
rate-limiting for the reaction occurring, while at the same time, maintaining thiosulfate additions as low as possible. That is,
S represents nitrate concentration, not thiosulfate concentration. Fill out the following table to indicate how an increase in
each of the variables in the left-hand column will affect the three process variables. Assume all other factors listed in the
left-hand column remain constant and also that X, the reactor total suspended solids concentration, remains constant. Use
(+) = increase, (−) = decrease, (0) = no change, and (i) = need more information to tell.
Process Variables
Variable Reactor Size, V (m3 ) Effluent Nitrate (mg/l) Thiosulfate Required (kg/d)
q̂
Q0
S0
θx
Xi0
13.38 A wastewater has a BODL of 1200 mg/l and a NO− 3 -N concentration of 400 mg/l. For denitrification in a CSTR with
recycle, operating at θ x = 10 d, will the electron donor or electron acceptor be rate-limiting to the overall reaction? Assume
ammonia-nitrogen is available and used for cell synthesis. Provide appropriate calculations to support your answer.
13.39 Draw a flow diagram for a two-stage organic oxidation, nitrification plant, designed in such a way as to reduce
overall oxygen requirements.
13.40 In a study of denitrification of an industrial wastewater, 120 mg/l acetate was used in the removal of 30 mg/l
nitrate-nitrogen. Estimate fs and fe.
13.41 What concentration of lactate would be required for denitrification of 60 mg/l nitrate-nitrogen in a CSTR with
settling and recycle in which θ x was maintained at 6 d?
13.42 Draw schematic diagrams of two economical biological treatment systems that you might design to achieve
removal of ammonia, nitrite, and nitrate-nitrogen from a municipal wastewater. For the first system (low nitrogen removal),
assume the requirement is simply to remove about 50% of the nitrogen, and in the second treatment system (high nitrogen
removal), assume the requirement is to remove 95% of the nitrogen. Indicate all f1uid and gaseous inputs and outputs from
each reactor, indicate what chemicals, if any, would be added to each reactor in the system, and indicate the dominant
forms of nitrogen in streams entering and exiting each reactor. Indicate what are the dominant electron donors and
acceptors for the biological reactions occurring in each reactor in the system.
13.43 A 10-m3 fluidized-bed biofilm reactor containing sand as the attachment surface is used for denitrification of a
wastewater containing 50 mg NO− 3 -N/l, and the total surface area within the reactor for the sand grains coated with a
8 2
growth of bacteria is 3(10 ) cm . Methanol is added as the primary substrate for bacterial growth, leading to the following
stoichiometric equation:
+
0.1667 CH3OH + 0.1343 NO−2 + 0.1343 H
= 0.0143 C5H7O2N + 0.06 N2 + 0.3505 H2O + 0.0952 CO2
a. If the reactor is assumed to be completely mixed, what minimum concentration of methanol in solution would be
required in order to ensure that an effluent nitrate-nitrogen concentration of 1.0 mg/l or less can be obtained?
b. Under such conditions, what would be the removal rate of nitrate-nitrogen in kilograms per day, and the detention time
for the reactor under these operating conditions (assume V equals the total reactor volume of 10 m3 )?
Assume deep biofilm kinetics apply (S w = 0). Pertinent coefficients are as follows:
Coefficient Units Methanol Nitrate-Nitrogen
D cm2/d 1.3 0.7
Df cm2/d 0.9 0.5
q̂ mg/mg VS a-d 9
K mg/cm 3 0.008 0.002
L cm 0.004 0.004
Xa mg VS a/cm3 15 15
b d−1 0.1 0.1

Nitrogen Transformation and Recovery

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Nitrogen Transformation and Recovery

Uploaded by

Copyright:

Available Formats

Source: Environmental Biotechnology: Principles and Applications, 2nd Edition

13. Nitrogen Transformation and Recovery

13.1. Nitrogen Forms, Effects, and Transformations

Table 13.1 Half-Reactions Associated with Biological Transformations of Nitrogen Species

Equation Description Reduction Equation

Source: Adapted from the supporting information in McCarty (2018).

Electron Donor Half-Reactions

Rd 1 Ammonium oxidation to nitrate

Rd 2 Ammonium oxidation to nitrite

Rd 4 Nitrite oxidation to nitrate

Rd 5 Organic oxidation of methanol

Electron Acceptor Half-Reactions

Ra1 Aerobic reactions

Ra2 Nitrate reduction to N 2

Ra3 Nitrite reduction to N 2

Rc1 Ammonium as N source

Rc2 Nitrate as N source

Rc3 Nitrite as N source

Energy Half-Reactions Synthesis Half-Reactions

Reaction Ra i Rdi Rc Rdj

Source: Adapted from the supporting information in McCarty (2018).

Aerobic organic oxidation Ra1 Rd 5 Rc1 Rd 5

Organic methanogenesis Ra4 Rd 5 Rc1 Rd 5

Heterotrophic denitrification with NO−

Heterotrophic denitrification with NO−

Autotrophic denitrification with NO−

Autotrophic denitrification with NO−

Anammox Ra3 Rd 3 Rc1 Rd 4

Aerobic organic oxidation skeleton equation:

Here, O2 has a molecular weight of 32 g, NH+

The other values can be estimated as outlined in Chapter 6.

Example 13.1: Analysis of a Reactor for Heterotrophic Denitrification

1 0.61 0.39 0.61 0.39

Thus, the concentration of BODL required for denitrification is

64 mg BODL 45 mg NO–3-N mmol NO–3-N 5 e meq NO–3-N

The concentration of ammonium-N required for biomass synthesis is

0.39(14) mg NH+ -N 45 mg NO–3-N mmol NO–3-N 5 e meq NO–3-N

The concentration of VSS formed is:

Example 13.2: Analysis of a Reactor for Autotrophic Denitrification

1 + 0.2 (0.02) (8)

Substituting these values in Equation 13.10 yields the skeleton reaction:

Thus, the concentration of H2 required in the influent flow for denitrification is

1 mg H2 45 mg NO–3-N mmol NO–3-N 1 e meq

The concentration of VSS formed is

Example 13.3: Nitrification in an Aerobic Bioreactor

0.66(32 g) 0.34(14 g) 0.34(113 g)

Then, considering BODL removal efficiency is 98%:

250 mg BODL 1 0.66(32 g O2)

250 mg BODL 1 0.34(113 g VSS)

0.34(14 g NH+ -N)

Substituting appropriate values into Equation 13.6, we obtain

1 0.084 0.916 1 0.084

Then, considering that ammonium removal efficiency is 95%:

Finally, the answers to the questions asked are

those nitrogen forms. Nitrification for removing NH+ 4

13.3.1. Biochemistry, Physiology, and Kinetics of Nitrifying Bacteria

NH2OH + 1.5 NAD+ = NO + 1.5 NADH + 1.5 H+

Nitric oxide is then oxidized to nitrate, capturing more energy in NADH:

NO + 0.5 NAD+ + 0.5 O2 + H2O = NO−

ΔG 0' = -45.44 kJ per e− eq

ΔG 0' = − 38.85 kcal per e− eq

Parameter Variations with Temperature

Parameter 5°C 10°C 15°C 20°C 25°C

fs0 0.14 0.14 0.14 0.14 0.14