EPA-R2-73-095
JANUARY 1973 Environmental Protection Technology Series
Development and Demonstration
of Nutrient Removal
from Animal Wastes
««<
Office of Research and Monitoring
U.S. Environmental Protection Agency
Washington. D.C. 20460
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and
Monitoring, Environmental Protection Agency, have
been grouped into five series. These five broad
categories were established to facilitate further
development and application of environmental
technology. Elimination of traditional grouping
was consciously planned to foster technology
transfer and a maximum interface in related
fields. The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL
PROTECTION TECHNOLOGY series. This series
describes research performed to develop and
demonstrate instrumentation, equipment and
methodology to repair or prevent environmental
degradation from point and non-point sources of
pollution. This work provides the new or improved
technology required for the control and treatment
of pollution sources to meet environmental quality
standards.
-------
EPA-R2-73-095
January 1973
DEVELOPMENT AND DEMONSTRATION OF
NUTRIENT REMOVAL FROM ANIMAL WASTES
R.C. Loehr, Professor
T.B.S. Prakasam, Research Associate
E.G. Srinath, Research Associate
Y.D. Joo, Research Assistant
Project #13040 DPA and 13040 DDG
Project Officer
Mr. Jeffery D. Denit
Environmental Protection Agency
Washington, D. C. 20460
Prepared for
OFFICE OF RESEARCH AND MONITORING
ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D. C. 20460
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, B.C. 20402
Price $3.45 domestic postpaid or $3 QFO Bookstore
-------
EPA Review Notice
This report has been reviewed by the Environmental Protection
Agency and approved for publication. Approval does not signify
that the contents necessarily reflect the views and policies
of the Environmental Protection Agency, nor does mention of
trade names or commercial products constitute endorsement or
recommendation for use.
-------
ABSTRACT
Laboratory and pilot plant studies evaluated processes applicable to
the removal of nitrogen, phosphorus, and color from animal wastewaters.
Three processes were evaluated: a) chemical precipitation of phos-
phorus, b) ammonia removal by aeration and c) nitrification followed
by denitrification.
Alum, lime and ferric chloride could be used to remove phosphorus and
color from animal wastewaters. The chemical costs to treat 1000 gallons
of poultry and dairy manure wastewaters were at least 10 times, and those
to treat duck wastewater were 2-3 times those quoted for phosphate
removal from municipal wastewaters. Two predictive relationships were
determined that appear useful for design and operation of possible
phosphate removal systems for these wastes.
Ammonia removal by aeration was found to be feasible from animal waste-
waters. Detailed equations were developed and verified to determine
the rate and amount of ammonia desorption under both quiescent and
aerated conditions. Ammonia desorption coefficients, KD, were found
to be a function of temperature, air flow rate, liquid volume, and
viscosity.
It was technically feasible to incorporate a nitrification-denitrifi-
cation sequence with the biological treatment of animal wastes to con-
trol the nitrogen content of the treated wastes. Nitrate formation was
inhibited by specific free ammonia concentrations. Approximately 65-75%
total nitrification represents the maximum amount of oxidized nitrogen
produced with poultry wastes. A minimum SRT value of 2 days was neces-
sary to sustain nitrification.
This report was submitted in fulfillment of project 13040 DPA and
13040 DDG between the Environmental Protection Agency and Cornell
University, Ithaca, New York.
m
-------
TABLE OF CONTENTS
TABLE OF CONTENTS v
LIST OF TABLES viii
LIST OF FIGURES xi
CONCLUSIONS 1
Phosphorus Removal 1
Ammonia Desorption 2
Nitrification-Denitrifi cation 4
RECOMMENDATIONS 7
Specific Recommendations 7
PROJECT NEED AND OBJECTIVES 9
Project Need 9
Project Objectives and General Results 12
Investigative Facilities 12
PHOSPHORUS REMOVAL 17
Introduction 17
Methods of Removal 17
General 17
Chemical Precipitation and Coagulation 19
Lime and Magnesium Treatment 20
Alum and Iron Treatment 21,
Coagulant Dose Determination 22
Objectives and Methods 23
Results 25
General 25
Concentration Effects 25
pH Effects 28
Predictive Unit Relationships 32
Other Parameters 48
Solids Production 49
Comparison of Chemicals 49
Cost Relationships 50
Significance of the Research 57
NITROGEN REMOVAL BY AMMONIA DESORPTION 61
Introduction 61
Theoretical Considerations 61
Effect of Dissociation 61
Transfer of Ammonia During Desorption 66
Ammonia Desorption from Animal Wastes 71
Objectives and Methods 72
Mathematical Approach 72
-------
TABLE OF CONTENTS continued.
Batch Desorption Systems 73
Continuous Desorption Systems 74
Variable pH 74
Aeration Towers 77
Materials and Methods Used in the Study 79
Materials 79
Methods 80
Experimental Setup 80
Results 85
General 85
Small Scale Studies 85
Aeration Tower 85
Diffused Aeration Studies 87
Quality of Effluent from Ammonia Stripping Systems .... 104
Large Scale Studies at the Pilot Plant 104
Predictive Relationships 113
General 113
pH, Temperature, and Air Flow 113
Effect of Rate of Aeration at Different Temperatures ... 117
Surface Tension and Viscosity 126
Significance of the Research 127
Combined Predictive Relationships 127
Place of Ammonia Desorption as a Method of Treatment ... 134
NITROGEN REMOVAL BY NITRIFICATION-DENITRIFICATION 141
Introduction 141
Methods of Removal 141
General 141
Practical Applications of Denitrification 142
Microbiology and Biochemistry of Nitrification 144
Formation of Ammonia 144
Nitrification 146
The Nitrifying Organisms 146
Physiology of the Nitrifying Organisms 147
Inhibitors of Nitrification 147
Biochemistry of Nitrification 148
Factors Affecting Nitrification 149
Denitrification 152
Chemical Denitrification 153
Microbial Denitrification 154
Denitrification in Waste Treatment Systems 156
Objectives and Methods 158
Objectives of the Study 158
Materials and Methods 159
Feed Suspension 159
Mineral Salts Solution 159
Continuous Flow Units 160
Batch Units - Nitrification 162
-------
TABLE OF CONTENTS continued.
Denitn'fication Reactors 162
Methods 164
Storage of Samples 165
Respirometry 165
Description of Terms 165
Results 167
Nitrification 167
Denitrification 209
Nitrification of the Denitrified Manure 245
Chemical Denitrification 248
Significance of the Research 250
General 250
Nitrification 250
Seed 250
Inhibition 251
Residual Ammonia 262
Residual TKN 266
Effect of Dilution 266
Effect of SRT, Loading Factor and pH 266
Denitrification 269
General 269
Anaerobiosis 269
Plateau in Denitrification 269
Denitrification Rates 269
Nitrogen and COD Removals 276
pH and Temperature 276
Gases Produced 276
Degree of Nitrogen Removal Possible in Poultry Manure. . . 277
Engineering Significance of the Research 277
Seed and Air Supply 277
Loading and SRT 277
pH and Temperature 279
Summary 280
Application of Nitrification-Denitrification Sequence to
Oxidation Ditch Operation 280
Cost Considerations 280
ACKNOWLEDGEMENTS 281
REFERENCES 283
APPENDIX 297
vn
-------
LIST OF TABLES
TABLE TITLE PAGE
1 Methods for Reducing the Phosphorus Content 18
of Wastewaters
2 Estimate of Chemical Requirements for Phosphate 30
Removal from Animal Wastewaters
3 Comparison of Chemicals 51
4 Characteristics of Animal Waste Slurries Assumed 53
for Chemical Cost Estimates in the Phosphorus
Removal Study
5 Cost Projections for Phosphate Removal from 54
Poultry Manure Wastewater
6 Cost Projections for Phosphate Removal from 55
Dairy Cattle Manure Wastewater
7 Summary of Ammonia Desorption Experiments 86
8 Correlations between 1C and Surface Tension 101
9 The Values of a and b for the Equation K = ay 103
10 Change in Waste Characteristics as a Result of 106
Ammonia Desorption
11 Values of M for the Equation A/LQ = M • KD 120
12 Values of m and 8 for the Equation M = me + e 122
u
13 Ratios of KD at Noted Temperatures to KD at 20°C 123
for Specific Rates of Air Flow
14 Values of the Temperature Coefficient E in the 123
Equation Kn = Kn E(e2"ei)
D2 Ul
15 Comparison of Experimental and Predicted 1C Values 125
16 Time Required to Obtain Specific Ammonia Removals 132
as Related to pH, Air Flow Rate, and Temperature
17 Relative Efficiencies of Ammonia Desorption 136
vm
-------
LIST OF TABLES continued.
TABLE TITLE PAGE
18 Performance of Continuous Flow Units with Solids 168
Recycling: Nitrification Study
19 Performance of Continuous Flow Units: 170
Nitrification Study
20 Summary of the Data of Performance of Unit E' at 171
Various SRT
21 Summary of the Data of Performance of Unit F1 at 172
Various SRT
22 Summary of the Data of Performance of Unit G1 at 173
Various SRT
23 Nitrification Results - Batch Study 181
24 Effect of pH on the Nitrification of Unit A1 186
Mixed Liquor
25 Effect of pH on the Nitrification of Oxidation 187
Ditch Mixed Liquor
26 Nitrogen Data of ODML at Various Dilutions - 191
Respirometric Study
27 Adjusted Nitrogen Data 191
28 Nitrogen Balance in a Highly Nitritifying 195
Oxidation Ditch Liquor - Respirometric Study
29 Nitrogen Balance in the Highly Nitratifying Unit A1 197
Mixed Liquor - Respirometric Study
30 BOD of Untreated and 0.1 M NH.C1 Treated Poultry 198
Manure Mixed Liquor
31 Effect of NH4-N on the Nitrification of Unit G1 200
Mixed Liquor - Respirometric Studies - Unadjusted
Uptake Data
32 Effect of SRT on the Tolerance of Ammonium 207
Chloride by Mixed Liquors
33 Effect of the Addition of Exogenous Hydrogen Donors 226
on Denitrification Rates of Poultry Waste
ix
-------
LIST OF TABLES continued.
TABLE TITLE PAGE
34 Dry Solids Content of Wet Manure Used in 232
Denitrification Study
35 Effect of Dilution and Supplemented N02-N, 236
NCL-N, and NH.-N on the Rate of Denitrification -
Run X
36 Relative Proportion of Some Gases Contained in 238
the Head Gases of the Denitrification Reactor -
Denitrification Run XI
37 Rates of Denitrification Observed in Reactors 241
Adjusted Initially to Various pH Values -
Denitrification Run XI
' 38 Free NH3~N and Nitratification in Batch and 258
Continuous Flow Units
39 Analyses of Materials and Percentages of Nitrogen 265
Mineralized on Incubation with Soil
40 Observed Rates of Denitrification 272
41 Nitrogen Removal Possible in a Process Involving 278
Two Sequences of Nitrification and Denitrification
-------
LIST OF FIGURES
TITLE PAGE
Floor Plan and Research Areas in the Agricultural 14
Waste Management Laboratory
2 Activities at the Agricultural Waste Management 15
Laboratory
3 Phosphorus Removal Patterns from Poultry Manure 26
Wastewater
4 Relationship of Initial Orthophosphate Concen- 27
tration on the Chemical Dosage
5 Removal Patterns as Affected by Initial Phosphate 29
Concentration - Poultry Wastewater
6 Total Phosphate Removal Using Lime and Alum 31
7 Effect of pH Control on Alum Dosage 33
8 Chemical Costs for Phosphorus Removal at 34
Controlled pH Values Using Alum
9 Percent Orthophosphate Removal Related to 36
Chemical Dose and Initial Alkalinity - Poultry
and Dairy Manure Wastewaters
10 Percent Orthophosphate Removal Related to Chemical 37
Dose and Initial Alkalinity - Duck Wastewater
11 Percent Orthophosphate Removal Related to Chemical 39
Dose and Initial Orthophosphate
12 Percent Phosphate Removal Related to Chemical 40
Dose and Remaining Phosphate - Poultry and Dairy
Manure Wastewaters
13 Percent Phosphate Removal Related to Chemical Dose 41
and Remaining Phosphate - Poultry, Dairy, and Duck
Wastewaters
14 Percent Orthophosphate Removal Related to 43
Chemical Dose and Initial Total Solids -
Duck Wastewater
15 Percent Orthophosphate Removal Related to 44
Chemical Dose and Initial Hardness - Poultry
and Dairy Manure Wastewaters
-------
LIST OF FIGURES continued.
FIGURE TITLE PAGE.
16 Lime Dose Related to Initial Alkalinity and 46
pH After Lime Addition
17 Orthophosphate Removal Related to pH and Initial 47
Orthophosphate Using Lime
18 Effect of Temperature on the lom'zation Constants 64
for Water and Ammonia
19 Effect of Temperature and pH on the Fraction of 65
Undissociated Ammonia
20 Graphical Procedure to Determine the Fraction 67
of Undissociated Ammonia
21 Schematic of the Transfer of Ammonia from a Liquid 69
to a Gas Phase
22 Reduction of pH During Ammonia Desorption 75
23 Mass Transfer During Countercurrent Air 78
and Liquid Flow
24 Equipment Used in Aeration Tower Experiments 81
for Ammonia Stripping
25 Schematic of Bench Scale - Batch Study Equipment 82
for Ammonia Desorption
26 Schematic Diagram of Continuous Flow Laboratory 83
Experiments for Ammonia Desorption
27 Pilot Plant Equipment for Batch and Continuous 84
Flow Ammonia Desorption Studies
28 Aeration Tower Experiments Poultry Wastewater 88
Different pH Levels
29 Aeration Tower Experiments Tap Water Plus NH.C1 89
Different pH Levels *
30 Aeration Tower Experiments Variation of KD with 90
Initial Ammonia Concentration and Liquid Flow Rate
31 Aeration Tower Experiments Variation of K with 91
Air Flow Rate and Air: Liquid Ratio D
xii
-------
LIST OF FIGURES continued.
TITLE PAGE
Diffused Aeration Experiments Tap Water Plus 93
NH4C1 Different pH Levels
33 Diffused Aeration Experiments Poultry Waste- 94
water Different pH Levels
34 Diffused Aeration Experiments Variation of KD 95
with Initial Ammonia Concentration
35 Reduction of Liquid Temperature During Ammonia 96
Desorption
36 Effect of Temperature and Air Flow on the Rate 97
of Desorption of Ammonia from its Solution in
Water
37 1C Variations Due to Temperature Under Quiescent 99
Conditions
38 Effect of Temperature on Surface Tension and 100
Viscosity of Suspensions of Poultry Waste in
Tap Water
39 Effect of Viscosity on Ammonia Desorption Rate 102
40 Solids Content of a Liquid as it Affects KD 105
41 Ammonia Variations During the Desorption of Fresh 108
Poultry Manure Wastewater
42 Change in the Ammonia Desorption Rate Due to 109
Changes in Viscosity
43 Effect of Air Flow and Viscosity on Kn - Large 110
Scale Studies D
44 Comparison of Kn Values from Laboratory and Pilot 112
Plant Studies u
45 Ammonia Removed Per Volume of Air as Affected by 115
pH and Temperature - Poultry Manure Wastewater
46 Ammonia Removed Per Quantity of Air - Dairy 116
Manure Wastewater
xm
-------
LIST OF FIGURES continued.
FIGURE TITLE PAGE
47 Quantity of Ammonia Removed Per Quantity of Air 118
Related to Temperature and Ammonia Concentration
48
The Variation of KV (Equation 45) with Temperature 119
49 Slope of the 1C - Air Flow Ratio as a Function of 122
Temperature
50 Desorption Time to Obtain Specific Ammonia 131
Removals as a Function of Temperature and pH
51 Equipment Used for Continuous Flow 161
Nitrification Studies
52 Schematic of Units Used for Batch Nitrification 163
and Denitrification Studies
53 TKN, Ammonia, and Percent Nitrification Laboratory 175
Continuous Flow Units
54 General Results from Nitrification Batch Unit "b" 176
55 Nitrogen Data from Nitrification of Batch Unit "b" 177
56 COD, Solids, and Total Nitrogen Nitrification of 178
Batch Unit 40
57 Nitrogen Data from Nitrification of Batch Unit 40 179
58 Maximum NH4-N, Nitrification and Initial TKN - 183
Batch Nitrification Units
59 Control of pH During Nitrification - Nitritifying 184
Mixed Liquor
60 Oxygen Uptake of ODML at Varying Initial pH Values 185
61 Oxygen Uptake of Diluted Mixed Liquor and Mixed 190
Liquor with Added Nitrogen - Nitrification Study
62 Oxygen Uptake of Diluted Mixed Liquor - Nitrification 194
Study
63 Oxygen Uptake of Diluted Nitrifying Mixed Liquors 196
xiv
-------
LIST OF FIGURES continued.
TITLE PAGE
Effect of Ammonia on the Oxygen Uptake Unit G1 201
Nitrification Study
65 Percent Nitrification Related to Ammonia and 203
TKN Concentrations
66 Effect of NH.C1 Concentrations on the Oxygen 204
Uptake of Mixed Liquor from Units E' and F1
67 Effect of Ammonia on the Oxygen Uptake of 206
Nitrifying Mixed Liquors
68 Effect of Ammonia on the Oxygen Uptake of Stored 208
Nitrifying Mixed Liquors
69 Inhibition of Nitrification with N-Serve 210
70 Denitrification of Unit A1 Mixed Liquor - 211
Run I
71 Denitrification at 20°C and 35°C - Run II 213
72 Denitrification at 20°C and 35°C with Manure 214
as an Added Hydrogen Donor - Run III
73 Denitrification at 20°C and 35°C with Engodenous 216
Hydrogen Donors from a Nitratified Mixed Liquor -
Run IV
74 Denitrification at 20°C and 35°C with Engogenous 217
Hydrogen Donors from a Nitritified Mixed Liquor -
Run V
75 COD Decrease Related to Oxidized Nitrogen 219
Decrease - Run V
76 Denitrification Due to Endogenous and Exogenous 221
Hydrogen Donors - Run VI
77 COD Reduction in Denitrification Due to Various 222
Hydrogen Donors - Run VI
78 Nitrogen Patterns in Denitrification Due to 224
Excessive Hydrogen Donors - Run VII
xv
-------
LIST OF FIGURES continued.
FIGURE TITLE
79 COD Removal Patterns with Different Exogenous 225
Substrates - Denitrification Unit - Run VII
80 Denitrification of Unit G1 Mixed Liquor with 228
Exogenous Hydrogen Donors - Run VIII
81 Theoretical and Observed COD Decreases - Denitri- 229
fication Run VIII
82 Denitrification at Various Wet Manure Loadings - 231
Denitrification Run IX
83 Denitrification of Diluted Nitritifying 233
Mixed Liquor - Run X
84 Denitrification of Diluted Nitritifying Mixed 234
Liquor Supplemented with Nitrogen - Run X
85 Denitrification of a Concentrated Nitritifying 235
Mixed Liquor - Run X
86 Denitrification of pH Controlled Mixed Liquors - 239
Run XI
87 Nitrogen Remaining in Reactors with Initial pH 240
Adjustment - Denitrification Run XI
88 pH Changes During Denitrification Head Gases - 242
Run XI
89 Typical Chromatogram of Denitrification 244
Head Gases - Run XI
90 Nitrification of Denitrified Poultry Wastes 246
Results of Reactors b and c
91 Nitrification of Denitrified Poultry Manure - 247
Nitrifying Organisms Added - Results of Reactors
a and d
92 Effect of Mercuric Chloride on Denitrification 249
93 Nitrite and Nitrate Formation as Affected 253
by the Free Ammonia/TVS Ratio
94 Nitrite and Nitrate Formation as Affected 254
by the Nitrous Acid/TVS Ratio
xvi
-------
LIST OF FIGURES continued.
FIGURE TITLE PAGE
95 Nitrate Formation Related to the Free 256
Ammonia Concentration
96 Nitrite Formation Related to the Nitrous 261
Acid Concentration
97 Oxidized Nitrogen Compounds Related to the 263
Ammonium Nitrogen Concentration in Continuous
Flow Nitrification Units
98 Nitrification Related to Solids Retention Time 267
99 Nitrification as Affected by the Loading 268
Factor - Comparison of Data from Various
Studies
100 Denitrification Data from Ref. 166 - 270
Interpretation of Results
xvii
-------
CONCLUSIONS
This report summarizes the results obtained in two projects - 13040 DPA -
"Tertiary Treatment of Animal Waste Waters" and 13040 DDG - "Animal
Waste Management - Demonstration of Feasible Handling and Treatment
Processes". The objectives of the two projects were similar, i.e., to
evaluate feasible processes applicable to the removal of nitrogen, phos-
phorus, and color from animal wastewaters and to indicate the manner
they may be integrated into animal waste management systems. Project
13040 DPA was a laboratory investigation to evaluate a number of possible
processes. The feasible results determined from the laboratory study
were demonstrated as part of 13040 DDG.
Three processes were evaluated by these projects: a) chemical precipi-
tation of phosphorus, b) ammonia removal by diffused and tower aeration,
and c) nitrification followed by denitrification. The conclusions from
the three phases of the project are summarized separately.
Phosphorus Removal
Three chemicals, alum, lime, and ferric chloride, were used to evaluate
the removal of phosphates from duck farm wastewater, poultry manure
wastewater, and dairy manure wastewaters.
1. When alum was used with all wastewaters and when high concentra-
tions of ferric chloride were used with dilute poultry manure waste-
waters, color removal was good and the color removal and phosphate
remaining curves were similar. The similarity suggests the possible
use of color as a routine operational tool for general process perfor-
mance when chemical precipitation is practiced with animal wastewaters.
When lime was used, color removal was poor.
2. Each wastewater had its own chemical demand. The chemical demand
of poultry manure wastewater was larger than for dairy manure and duck
wastewater respectively. The chemical demand appeared to be in pro-
portion to the phosphate concentration in the wastewaters.
3. The concentrations of alum necessary to obtain a specific percent
phosphate removal from poultry and dairy manure wastewater were less
than those required for lime. With duck wastewater required lime
dosages were less than those for alum.
4. The quantity of lime required to raise the pH of poultry and dairy
manure wastewaters to about 11 was about 1.0 to 1.1 times the alkalinity
in the wastewaters. The quantity of lime required to raise the pH to
11 for duck wastewater varied from about 1.5 to 3.0 times the alkalinity
of the wastewaters.
5. When lime was used, high phosphate removals were not obtained at
-------
high pH levels. For specific wastewaters, the percent removals for
wastes having initial orthophosphate concentrations above 100 mg/1
were reasonably consistent irrespective of concentration. Considerable
removals, 60-80%, were obtained with pH control in the range of 9.0
to 9.5.
6. A number of relationships were explored to relate reduction of
phosphate to chemical dosages if the waste characteristics are known.
The most sensitive parameters for all wastewaters were: a) the chemical
dosage per remaining total or orthophosphate concentration versus per-
cent total or orthophosphate removal, and b) chemical dosage per initial
calcium or total hardness versus total or orthophosphate removal. Both
parameters appear useful for design and operation of possible phosphate
removal systems. The parameters were reasonably sensitive for data
from different wastewaters, for different phosphate concentrations, and
for the three chemicals investigated.
7. The costs of chemical precipitation of phosphorus from poultry and
dairy manure wastewater were estimated using "average" wastes. Lime
was found to be the least expensive chemical due to lower chemical costs.
Chemical costs to treat 1000 gallons of these wastewaters were at least
10 times or more than those quoted for phosphate removal from municipal
wastewaters. The chemical costs of removing phosphates from duck waste-
water were lower but were still 2-3 times those quoted for municipal
wastewater.
The chemical costs per animal over the total phosphate range of 50 to
95% ranged from $0.28 to $1.00 per day per 100 dairy cattle and $0.17
to $0.40 per day per 1000 laying hens. For duck wastewater, the chemi-
cal costs over an orthophosphate removal of 50-90% ranged from 0.7-4.3
cents/1000 gallons of wastewater/day for lime, 2.2-4.3 cents/1000
gallons/day for alum and 9.1-25.5 cents/1000 gallons/day for ferric
chloride.
Ammonia Desorption
Diffused aeration and aeration towers were used to evaluate the feasi-
bility of this process with poultry and dairy manure wastewaters.
Desorption coefficients found in such laboratory and pilot plant experi-
ments are system specific, being influenced by such factors as system
geometry, type of diffuser and tower systems used, and mixing and tur-
bulence in the systems. Care should be taken in utilizing the desorption
coefficients determined in these experiments to other situations and in
extrapolating the equations beyond the conditions used in the experiments.
8. Detailed equations have been developed which were used to determine
the ammonia desorption coefficient, 1C, under both quiescent and aerated
conditions. Additional equations were developed to determine the
ammonia lost under a variety of environmental conditions. These equations
were verified in both laboratory and pilot plant scale experiments and
-------
shown to be valid at both levels. KD values obtained from laboratory
units were found to be close estimates of the values that occurred in
larger systems. The K- values determined in these studies ranged from
0.004 to 0.045 per hour depending on process conditions.
9. At a given temperature, higher values of 1C can be obtained by
increasing the rate of aeration. 1C increases with an increase in air
flow rate and decreases as the liquid volume of the desorption unit
increases.
10. The value of 1C is a function of the temperature of the liquid.
Over the range of 10-35°C the values of KD are increased by a factor of
1.5 to 2 when the temperature is raised by 10°C. An empirical equation
can be used to predict changes in KD as the air flow and temperature
change.
11. The pH of the system also will affect ammonia desorption since it
affects the free ammonia available for desorption. At a pH of 8-9, a
5°C temperature decrease doubled the desorption time to achieve a
desired removal. At a pH of 10-11, a 10°C temperature decrease doubled
the time necessary to achieve a specific removal. The larger tempera-
ture effect at the lower pH values occurred due to the cumulative effect
of pH on the free ammonia concentration and the effect of temperature on
V
12. Little difference in desorption time occurred at pH values of 10-11
at temperatures from 10°C to 30°C. Below a pH of 10, the time to accom-
plish a specific removal increased rapidly. Higher removal efficiencies
increased desorption time. The time to accomplish 75% removal was
double that to accomplish 50% removal at 20°C. In the same manner, it
took twice as much time to accomplish 99% removal as to accomplish 90%
removal.
13. Viscosity was found to affect the rate of desorption, 1C.. As
viscosity increased, 1C decreased.
14. The equations developed in this phase of the study appear able to
be used to estimate the magnitude of design parameters to remove ammonia
from a specific waste. Many factors affect the design of ammonia
desorption systems such as pH, temperature, air flow rate, liquid volume,
time of desorption, and the characteristics of the waste. The equations
that were developed in this project may be used to evaluate the inter-
relationships of these variables to investigate the feasibility of
specific ammonia desorption units.
15. The results of the study permit better estimates to be made of the
-------
ammonia loss and the nitrogen balances in systems such as aerated bio-
logical treatment systems, swift streams, impoundments, aerated odor
control systems, waste storage units, and animal confinement units.
With the amount of ammonia that is volatilized better estimated, its
effect on the local environment may be better determined.
Nitrification-Denitrification
Nitrification studies were done in batch and continuous flow units and
in respirometric experiments. Denitrification experiments were con-
ducted in batch units. Poultry manure wastewaters were used in the
experiments.
16. Nitrate formation was inhibited by free ammonia concentrations in
excess of 0.02 to 0.033 mg/1 and at ratios of free ammonia to total
volatile solids greater than 0.000025. Nitrite formation did not
appear inhibited by nitrous acid concentrations up to 4 mg/1. Nitrite
production predominated in units having a high loading rate.
17. Approximately 50-60% of the initial TKN in poultry manure waste-
water could be nitrified in a biological nitrification process. Con-
siderable ammonia nitrogen remained in the nitrification unit. When
these nitrified wastes were denitrified, subsequent nitrification of
the denitrified waste resulted in nitrification of the residual ammonia.
This second stage nitrification amounted to an additional 15% nitrifi-
cation of the initial TKN. Thus the total nitrification that was
accomplished in the two stages (50-60% + 15%) represented the maximum
amount of oxidized nitrogen that can be produced with these wastes.
18. A minimum SRT value of 2 days was necessary to sustain nitrification
in biological treatment units. Growth kinetics obtained in other studies
appeared applicable to poultry manure.
19. Loading factors less than 0.8# COD/day/#MLVSS insured maximum nitrifi
cation. At factors greater than 0.15, nitrites predominated.
20. Nitrification units functioning with poultry wastes maintained
nitrification even at a pH of 4.9. Adjusting the pH to neutral or
alkaline conditions did not cause any significant increase in nitri-
fication and there was no apparent need to control the pH in the nitri-
fication of poultry manure.
21. Denitrification of nitrified poultry manure followed a phasic
pattern with a plateau occurring after an initial rapid rate of removal.
Another rapid rate followed the plateau.
22. Denitrification could be caused by using either endogenous or
exogenous hydrogen donors. The appropriate exogenous donors are those
that do not increase the total nitrogen in the system.
-------
23. Denitritification (reduction of nitrite) rates were greater than
denitratification (reduction of nitrates) rates increasing the advan-
tages of a heavily loaded system in which nitrites predominate for
nitrogen control.
24. The gases resulting from denitrification of the nitrified poultry
manure were determined to be nitrous oxide, O; nitrogen gas, N2;
nitric oxide, NO; and carbon dioxide, CCL. Nitrogen gas appeared to
be the major end product.
25. It is technically feasible to incorporate a nitrification-denitri-
fication sequence with the biological treatment of animal wastes to
control the nitrogen content of these wastes.
-------
RECOMMENDATIONS
Summary - From the results and conclusions of this investigation, it is
clear that chemical precipitation systems can remove color and phos-
phorus from animal wastewaters, that ammonia desorption systems can be
utilized for nitrogen removal, and that it is possible to incorporate
a nitrification-denitrification sequence in an aerobic treatment process
for animal wastewaters. The practical applicability of these processes
to the treatment and disposal of animal wastes, however, is not clear
in all cases.
Although one phase of this project was directed toward chemical means
of removing phosphates from animal wastewaters, it should not be inferred
that this approach is the most effective method of phosphate control
for these wastewaters. The required chemical concentrations needed are
in proportion to characteristics of the wastewater such as alkalinity,
hardness, or phosphate. Ratios of chemical dosage per initial ortho-
phosphate concentration ranged up to 8-10 for alum and lime at low,
residual orthophosphate concentrations (less than 5-10 mg/1) and at
greater than 90% orthophosphate removals. Sludge production represents
subsequent handling and disposal problems and ranged between 0.5-1.0 mg/1
suspended solids increase per mg/1 of chemical used.
To achieve low residual phosphate concentrations, a wastewater containing
100 mg/1 of orthophosphate may require about 800-1000 mg/1 of chemicals
which may produce an additional 400-1000 mg/1 of suspended solids for
ultimate disposal. The large chemical demand, the sludge production,
and the additional cost are disadvantages to this method of phosphate
control for concentrated animal wastewaters. Approaches other than
conventional liquid waste treatment methods are needed for animal waste
phosphate control.
Ammonia removal from animal wastewaters by desorption in aeration systems
will occur but is not likely to be an ultimate solution to the nitrogen
control problem with these wastes. Ammonia desorption transforms the
nitrogen problem from a liquid to a gaseous concern. The gaseous
ammonia in the atmosphere will contribute to the nitrogen contribution
of precipitation which again enters the surface waters. Nitrification-
denitrifi cation offers the opportunity to transform a liquid nitrogen
problem into an innocuous gaseous form, nitrogen gas, without subsequent
potential environmental quality problems.
Specific Recommendations
1. Controlled land disposal should be considered a high priority method
for phosphorus control from agricultural wastes and wastewaters rather
than chemical precipitation methods. Land disposal is more amenable to
normal agricultural operations, avoids the need for chemical control
and treatment plant operation, and eliminates additional problems of
-------
chemical costs and chemical sludge production, handling, and disposal.
2. The predictive relationships observed in this study for phosphorus
removal should be closely explored with other wastes to determine their
general applicability. These relationships were reasonably sensitive
for data from a number of different wastewaters, for different phos-
phate concentrations, and for the three chemicals investigated.
3. Controlled land disposal using waste application rates that will
not result in nitrogen leaching and runoff should be utilized with
animal wastes as a positive method to avoid environmental problems
from animal production operations.
4. Detailed controlled field investigations are needed to delineate
nitrogen loading rates that can be applied to specific land types with-
out causing leaching and runoff problems. Such studies need to be done
on a large scale using treated and untreated wastes to establish soil
assimilation capacities for different soils, geographical locations,
and land management opportunities.
5. Where the nitrogen content of animal wastewaters needscontrol before
disposal, a nitrification-denitrification process should be utilized in
preference to ammonia desorption.
6. More detailed laboratory and field scale studies are necessary to
establish the design parameters for nitrification and denitrification
in animal waste treatment systems. A nitrification-denitrification
process has the multiple benefits of odor control, BOD and solids
reduction, as well as nitrogen control.
7. The ammonia desorption relationships in aerated systems that were
developed in this study can be used to account for nitrogen losses in
natural and aerated systems, to develop better nitrogen balances on
treatment systems, and to determine the effect of environmental changes
on the efficiency of ammonia desorption systems. Detailed studies
should be undertaken to evaluate these relationships and their impli-
cations to other waste treatment and natural systems.
-------
PROJECT NEED AND OBJECTIVES
PROJECT NEED
Until recently agriculture had not been considered a serious source of
environmental contamination due to the diverse nature of its activities
and the comparatively small scale of each production unit. Located in
relative isolation surrounded by apparently unlimited land, any air and
water contaminants generated by agricultural operations usually were
sufficiently diluted or stabilized before potential problems became real.
Modern developments in agriculture and growing public use and develop-
ment of rural lands tend to eliminate this relative isolation.
Farm size and productivity per farm worker have increased significantly
in recent years. Intensive crop and livestock production have taken on
many aspects of industrial operations. The increased efficiency of
agricultural production has generated or has been associated with a
variety of environmental problems.
The specific role of agriculture in the environmental quality problems
of the nations remains unclear. Available information suggests that the
agriculturally caused pollution problems may be significant especially
at the regional and local level. Data on fish kills from feedlot run-
off, nutrient problems due to runoff from cultivated lands, the quan-
tities of animal and food processing wastes produced nationally, the
pollutional potential of these wastes, the possible contamination of
ground waters from crop production and land disposal of wastes, nuisances
due to odors and dust, and the increasing size of agricultural production
operations indicate that considerable attention must be given to the
development of proper methods to handle, treat, and dispose of agricul-
tural wastes with minimum contamination of the environment.
In response to the need for increased animal production, changes have
taken place in livestock production, feeding, slaughtering, transporta-
tion, and processing operations. These changes have resulted in con-
finement feeding of livestock and in increased numbers of animals per
production unit. Meat, milk, and eggs are produced increasingly in
confined large industrial type facilities. The trend toward controlled
and enclosed facilities, which is virtually complete for smaller animals
such as hens, broilers, and ducks, is increasing even for the larger
animals such as beef and dairy cattle. Commercial egg production is
almost 100% from confinement poultry houses. Present poultry manage-
ment permits the concentration of egg-laying hens in flocks of several
hundred thousand birds on a site consisting of small acreage. Swine
production also is moving toward large, confinement operations. The
trend in milk production is toward large herds and centralized large
bulk milk processing.
The amount of animal wastes that cause environmental quality problems is
-------
not well documented. It is incorrect to use the amount of waste defe-
cated by an animal to indicate the actual surface and ground water
pollution that may result. Only a small proportion of the wastes from
livestock operations may find its way into surface and ground water.
However, under conditions of intensified animal production, large
quantities of wastes must be disposed of under circumstances that can
cause problems comparable to the discharge of untreated municipal or
industrial wastes.
Historically, animal wastes have been recycled through the soil environ-
ment with a minimum of direct release to the water environment. With
increasing concentrations of livestock and alternative sources of fer-
tilizers, the practice of distributing the manure on the land has become
doubtful from a profit standpoint. Potential and actual runoff resulting
from the land disposal of wastes has caused examination of other approaches,
These alternative approaches have not yet solved the animal waste dis-
posal problem. If the wastes are fluidized, a water carriage and treat-
ment system is needed. Current aerobic and anaerobic microbial treat-
ment systems increase costs with only minor reduction in the total volume
of material to be handled. Dehydration, drying, or incineration
increases equipment and power costs and require air pollution control
measures.
There is no one type of treatment system that will be satisfactory for
every type of animal production facility. However, liquid waste treat-
ment systems are becoming more common. Anaerobic and aerobic lagoons,
oxidation ditches, and anaerobic digestion units can reduce the oxygen
demanding materials (BOD, COD) and the suspended solids content of the
untreated animal wastes but remove very little of the nitrogen and
phosphorus in the wastewater. In addition, the effluent from these
facilities will have an undesirable color which, depending upon the
amount of dilution in the receiving stream, may be a water quality
problem. Effluents with these characteristics may contribute to unde-
sirable nitrate concentrations in ground water.
The quantity of animal wastes produced in many states can exceed that
produced by municipal and industrial sources in terms of oxygen demanding
materials, solids, and nutrients. An estimate of the nitrogen and phos-
phorus contributions from various sources indicates that the contribu-
tions from domestic wastes are about 1500 million pounds of nitrogen
and 500 million pounds of phosphorus per year. The contributions from
industrial waste were estimated at greater than 1000 million pounds of
nitrogen per year.
An estimate of the nitrogen contribution from farm animals indicates a
contribution of almost 15,000 million pounds per year. Much of this
material is deposited on pasture lands and does not reach the receiving
waters. At the present time, about 25% of the nitrogen from farm
animals results from confined animal feeding operations. With the
increasing trend toward greater numbers of larger confinement production
10
-------
facilities, more of this contribution may reach the water environment
and may require control.
Many states are setting stream standards that do not permit ammonia con-
centrations greater than about 1 or 2 mg/1 and phosphorus concentrations
greater than 0.1 or 0.2 mg/1. High nitrogen and phosphorus concentra-
tions can be present in animal wastewaters. In order to meet current
stream standards, control of the nutrients in animal wastewaters will
be necessary in both point source discharges and land runoff.
Most of the untreated and treated animal wastes are disposed of on
nearby land. The organics and some of the nitrogen and phosphorus are
incorporated into crops and into the top layer of the soil. The
remainder of the nutrients will migrate toward the ground and surface
waters, the rate being dependent upon the quantity of nutrients and
liquid applied and the rainfall. Almost all of the phosphorus will be
removed from the liquid by soil particles within the top few feet of
soil. The reduced forms of nitrogen can be oxidized, ultimately to
nitrates, excess quantities of which can cause the rejection of waters
as a potable source for both humans and animals.
Where land disposal is an integral part of animal waste treatment opera-
tions, the concentration of nitrogen in the waste is likely to be more
critical than that of phosphorus. Processes for the removal of nitrogen
from animal waste treatment operations may be more important than those
removing phosphorus in these cases.
Where discharge to surface waters is practiced, the concentration of
both nitrogen and phosphorus in animal wastes may be critical. Poten-
tial processes for the removal of these nutrients are included in this
research.
Even where land disposal of agricultural wastes is practiced, there will
be the need to control application rates to levels that will not cause
secondary environmental quality problems such as contamination of
ground water, surface water by runoff, or the soil. Considerably more
information is necessary to delineate proper soil loading rates for
wastes of different characteristics and for different geographical
locations.
In earlier times, agricultural wastes were ignored by regulatory
agencies dealing with water pollution. As the degree of treatment
required of municipal and industrial wastes steadily increases and as
the specialization of agricultural operations increases, more attention
must be given to the abatement of pollution from agricultural sources.
Knowledge concerning the control of pollution from these sources is
important in formulating overall waste control, treatment, and disposal
requirements for agricultural, industrial, and municipal waste sources
and in formulating the water resource policies of the nation.
11
-------
PROJECT OBJECTIVES AND GENERAL RESULTS
The challenge is to develop processes that can remove undesirable com-
ponents from waste water as economically as possible. The challenge
becomes much larger when one contemplates processes for the treatment
of animal wastewaters. Systems treating these waters can expect: a)
concentrated, intermittent loads, b) to be operated in the open and
have wide seasonal temperature variations, and c) little maintenance
or skillful operation.
This report summarizes the results obtained in two projects: 13040 DPA -
"Tertiary Treatment of Animal Waste Waters" and 13040 DDG - "Animal
Waste Management - Demonstration of Feasible Handling and Treatment
Processes". The objectives of the two projects were similar, i.e., to
evaluate feasible processes applicable to the removal of nitrogen,
phosphorus, and color from animal wastewaters and to indicate the most
feasible processes and the manner they may be integrated to treat
animal wastewaters. Project 13040 DPA was a laboratory investigation
to evaluate a number of possible processes. The feasible results deter-
mined from the laboratory study were demonstrated as part of Project
13040 DDG.
These investigations evaluated three processes: a) chemical precipi-
tation of phosphorus, b) ammonia stripping, and c) nitrification-
denitrification as they can be applied to animal wastewaters, especially
dairy and poultry wastewaters. Detailed results of these three phases
are discussed in this report.
INVESTIGATIVE FACILITIES
The laboratory investigations were conducted in the agricultural waste
research laboratory located in Riley-Robb Hall, Cornell University.
This research laboratory comprises about 2000 square feet of space and
includes equipment for conducting the analyses pertinent to waste manage-
ment research as well as equipment for conducting biological, physical,
and chemical waste treatment studies.
The demonstration or pilot plant investigations were conducted in the
Agricultural Waste Management Laboratory on Game Farm Road, Ithaca,
New York. The latter facility was constructed with funds from Project
13040 DDG and consists of a building 60' x 136' containing tanks, pumps,
mixers, aeration units, process monitoring equipment, drying units,
animal housing arrangements, and all necessary utilities required for
conducting the demonstration research at the facility. The Laboratory
is unique in its mission and possibilities. Research at the Laboratory
can be accomplished with many animal wastes, on various waste manage-
ment processes, and in a variety of process flow schemes. The staff
working on projects in the Laboratory include over a dozen individuals
from a number of disciplines. Graduate M.S. and Ph.D. students use the
facilities in the Laboratory for their research.
12
-------
The Laboratory was designed to demonstrate feasible waste management
processes on a scale large enough to illustrate their application to
full scale animal production operations and to provide reasonable
estimates of the cost of installing and operating the processes. It
complements the basic research engineering activities in this waste
management area at Cornell and permits larger scale evaluation of
processes estimated to be feasible from basic studies. Approximately
5000 sq. ft. of space is used for pilot plant scale research for liquid
and solid animal wastes. An additional 2000 sq. ft. is used to study
the interrelationships of handling and treating animal wastes.
This facility is being utilized for a number of projects dealing with
pollution control from animal wastes in addition to the project described
in this report. A summary of the demonstration activities underway in
the Laboratory includes:
-evaluation of an oxidation ditch for poultry waste treatment
-evaluation of liquid versus "dry" waste management systems for poultry
wastes
-studies on drying of wastes
-investigation of ammonia loss by aeration and nitrification-denitri-
fication processes
-odor control by drying and aeration
-evaluation of microbiological contamination of confined animal pro-
duction operations.
These demonstration activities are being continued and initiation of
others is underway. Research supported by a variety of sources, i.e.,
Agricultural Research Service - USDA, EPA Traineeships, College of
Agriculture and Life Sciences, and Agway, Inc., as well as by these
EPA Demonstration Grants are conducted at the Laboratory. Thus this
facility serves as a center for broad research activities in agricul-
tural waste management.
The Laboratory also serves an extension and broad educational function.
Many animal producers interested in learning of better waste management
approaches, students and faculty from Cornell and other institutions,
personnel of local, state, and federal agencies, and interested urban
dwellers continually visit the Laboratory. These tours and visits
enhance the transfer of available technology in animal waste management
and are an important facet of the Laboratory activities.
Illustrations of the Laboratory and its activities are presented in
Figures 1 and 2.
13
-------
J
LABORATORY
-v/-
PILOT PLANT
PROCESS EQUIPMENT
POULTRY
HOUSING AND
WASTE
MANAGEMENT
WASTE STORAGE
FLOOR PLAN
PILOT PLANT
PROCESS EQUIPMENT
ANALYTICAL LABORATORY
FIGURE 1
FLOOR PLAN AND RESEARCH AREAS
IN THE AGRICULTURAL WASTE
MANAGEMENT LABORATORY
14
-------
RESEARCH AT THE AGRICULTURAL WASTE
MANAGEMENT LABORATORY
liun t!tf aims
VISITORS TO THE LABORATORY
FIGURE 2
ACTIVITIES AT THE AGRICULTURAL WASTE
MANAGEMENT LABORATORY
15
-------
PHOSPHORUS REMOVAL
INTRODUCTION
Our knowledge of the fundamentals and of appropriate engineering methods
for phosphorus removal primarily has come from studies on municipal
wastewaters. Although phosphorus removal also will be required of
industrial and other concentrated wastewaters, few studies have been
undertaken to determine if the basic information obtained from munici-
pal wastewaters can be extrapolated to concentrated wastewaters that
may have different characteristics. Such evaluations are needed if
industrial organizations are to prepare to meet increasingly stringent
effluent requirements and to develop optimum treatment facilities for
their wastes. Requirements are being set on both the percentage of
phosphorus removed from a wastewater and the actual concentration in
the effluent. The former requirement is likely to be less restrictive
for concentrated wastewaters. Information on appropriate treatment
strategies is needed to meet both requirements.
The principal benefit from removing phosphorus from wastewater is the
control of aquatic growths. Phosphorus is an essential nutrient which
can be limited to restrict algal growth. If phosphorus concentrations
in water are below 0.01 to 0.03 mg/1, algal growth may be limited.
Even at 0.3 to 0.5 mg/1, growth has been reported to be minimal. How-
ever, at progressively higher concentrations there is proportionately
more growth until nuisance conditions are reached. Excessive algal
growth can cause taste and odor in water supplies, deplete oxygen upon
death, become unsightly, increase water treatment costs, and create
other problems.
Management of both wastewater and receiving waters has been employed
to prevent excessive blooms of aquatic plants. Such treatment has
included the periodic application of algicides, diversion of nutrient
rich wastes to less sensitive or less valuable receiving waters, or a
combination of similar measures. Such management procedures have
limitations and do not attack the real causes of the problem. A more
positive approach is to remove the nutrients at their source.
METHODS OF REMOVAL
General - The principal methods for removing phosphorus from wastewater
include chemical precipitation, adsorption on chemical and on biological
floe, ion exchange, and chemical-biological treatment. A summary of
the possible methods are noted in Table 1.
The most feasible methods of phosphorus removal are biological, chemi-
cal, or the combination of chemical-biological processes. All three of
these approaches are directed toward converting soluble and colloidal
phosphorus into recoverable insoluble material. Chemical coagulation
17
-------
TABLE 1
METHODS FOR REDUCING THE
PHOSPHORUS CONTENT OF WASTEWATERS
Classification Method
Physical and Chemical Methods Land Application
Electrochemical
Ion Exchange
Precipitation
Sorption
Biological Methods Activated Sludge
Algal Utilization
has received the greatest attention and several effective chemical treat-
ment methods (1-4) have been proposed. Most of these chemical treatment
methods have been applied to the removal of phosphate from sewage treat-
ment plant effluents. In other investigations, iron and aluminum salts
have been added directly into the aeration tank of an activated sludge
system (5). Interest continues on the use of biological methods for
achieving adequate phosphate removals without adding another stage of
treatment.
The application of phosphorus removal methods to agricultural wastewater
becomes critical when the wastewaters are discharged to surface waters.
However, the phosphorus problem is minimized when wastewaters are dis-
charged to the land. Soil is a composite medium containing inert rock,
gravel, and sand as well as reactive clays and minerals. Varying
amounts of organic matter such as living and dead vegetable and animal
matter and humus are also a part of soil. Soil has the ability to
accomplish effective waste treatment by such mechanisms as biological
oxidation, chemical precipitation, ion-exchange, adsorption, chemical
oxidation, and nutrient uptake by plants. Many factors affect the
ability of soil to remove phosphorus, such as pH, particle size, type
of soil, temperature, organic content, oxidation-reduction potential, and
reaction time. Available data from a number of studies indicate that
most phosphate removal occurs in the top layer of soil. In most soils,
the infiltration capacity of a soil is likely to fail before the phos-'
phate removal capacity has been exceeded.
The potential of soil to treat wastes and to remove phosphorus should
not be ignored although relationships are not available to make general
predictions about the ability of a soil to act as a waste treatment
device. Advice of a competent soil scientist is valuable in assessing
the specific use of soils. When land is available, soil should be
18
-------
considered as a disposal medium for agricultural wastes and wastewaters
as well as sludges from municipal treatment plants.
There are, however, locations and times when the land is not adequate to
serve as a waste disposal medium. Under these conditions, more conven-
tional forms of phosphorus removal are necessary. The characteristics
of agricultural wastes are such that some form of biological treatment
will be necessary if discharge to surface waters is contemplated. When
required, phosphorus removal would be an additional step in the treatment
process. Chemical precipitation methods offer the greatest possibility
to provide the necessary removals.
Chemical Precipitation and Coagulation - The earliest phosphorus removal
investigations centered around the addition of various chemicals to
treated wastewater in order to form insoluble phosphorus compounds
which were removed by settling and/or filtration (1, 4). Chemical pre-
cipitation of phosphorus has progressed to the point where efficiency
and cost can be predicted with reasonable precision, at least for muni-
cipal wastewaters. Several municipal wastewater treatment plants are
now in operation, and others are being planned and built which employ
some form of chemical treatment for the removal of phosphorus from
wastewaters.
Owen (1) appears to be the first to report on the use of chemicals as a
tertiary step for the removal of phosphorus. He reported that the
addition of lime to secondary effluent, followed by mixing, flocculation,
and settling, reduced the effluent phosphorus concentration by greater
than 99%. The sludge produced in this operation was approximately three
times the volume ordinarily handled at the plant and was difficult to
dewater. The chemical requirements were quite high, 720 mg/1 as Ca(OH)?,
These difficulties have led investigators to evaluate the usefulness of
other chemicals for this purpose. In addition to lime, chemicals such
as alum, ferric sulfate, ferric chloride, ferrous sulfate, ferrous
chloride, and sodium aluminate have been used successfully for removing
phosphorus from wastewaters. In some instances organic polymers, used
in conjunction with the above chemicals, have proven useful. Precipi-
tated phosphates of calcium, iron, and aluminum frequently are colloidal
and are settled only after adequate coagulation. The selection of the
most economical chemical is dependent upon the wastewater characteristics,
the plant location, the desired efficiency, and the sludge handling and
disposal methods to be employed.
When comparing methods for phosphorus removal, the matter of handling and
the ultimate disposal of sludge should receive close evaluation. Chemical
methods for phosphorus removal produce large quantities of sludge. The
amount of sludge production depends upon the characteristics of the waste-
waters, the required degree of phosphorus removal, and the chemical used.
The problems associated with high chemical dosages and sludge handling
19
-------
and dewatering can be partially offset by chemical recovery processes.
It is not foreseeable that chemical recovery processes will be feasible
with agricultural wastewaters. These wastewaters are of higher strength
and lower volume than municipal wastewaters and are at locations where
sludge incineration or other chemical recovery methods are not likely
to be practical. In such cases, lagooning is likely to be the more
common method of chemical sludge disposal.
Lime and Magnesium Treatment - Lime reacts with the bicarbonate alka-
linity of wastewater to form calcium carbonate and also reacts with
orthophosphate to precipitate hydroxyl apatite:
Ca(OH)2 + Ca(HC03)2 -> 2CaC03 4- + 2H20 (1)
5Ca++ + 40H" + 3HPO + CaOH(P0) 4- + 3H0 (2)
The apatite precipitate, represented by Ca5(OH)(P04)3 in the above
equation, is a crystalline precipitate of variable composition. The
Ca/P mole ratio may vary from 1.3 to 2.0.
The reaction in Equation 1 is complete above a pH of 9.5 while that of
Equation 2 starts above a pH of 7 but is very slow below a pH of 9.
Phosphate removal improves as the pH is raised. As the pH increases
above 9.5, precipitation of magnesium hydroxide begins:
Mg++ + Ca(OH)2 + Mg(OH)2 4- + Ca++ (3)
Magnesium precipitation will not be complete until the pH reaches 11.
The solubility of hydroxylapatite decreases rapidly with increasing pH
with the result that phosphate removal improves with increasing pH.
In municipal wastewaters, it has been shown that essentially all ortho-
phosphate is converted to an insoluble form at pH values above 9.5.
Although the orthophosphate may be precipitated by calcium ions, poly-
phosphate is not readily removed unless orthophosphate is also present
so that the polyphosphate is adsorbed on the floe resulting from the
precipitation of the orthophosphate (6). In raw wastes, a significant
portion of the total phosphate will be organic phosphate and poly-
phosphates which are more difficult to remove with lime than the ortho-
phosphates. Biologically treated wastewater will result in the conversion
of the organic and polyphosphates to the orthophosphate form. Over 90%
of the total phosphate in biologically treated effluents can exist as
orthophosphates.
20
-------
The actual chemical dosage required to meet a specific phosphate resid-
ual will depend on the chemical demand of the wastewater which is a
function of many factors. Attempts to calculate the desired lime
dosage from stoichiometric relationships involving alkalinity, phos-
phate, calcium, and hardness analyses of the wastewater result in low
dosages. Empirical relationships, based on the use of jar or pilot
plant tests, are used to establish design and operating chemical
dosages.
When low residual phosphate concentrations are sought, lime may be added
to raise the pH of the wastewater to 11-11.5. With municipal wastewaters,
the lime requirement, to reach a pH of 11, has been indicated as about
1.5 to 2.8 times the alkalinity of the waste (7). The use of lime to
remove phosphates has been modeled (8) and the chemistry and precipi-
tation mechanisms summarized in detail.
Although calcium is the most common divalent ion used for phosphorus
precipitation, magnesium also has been used for this purpose (9). Solu-
ble salts of magnesium, such as magnesium oxide, hydroxide, or carbonate
have been utilized. When these salts were used, a mixture of MgNH.PO.-6O
and Mgo(PO.)p-4H?0 precipitated from a waste deficient in calcium. When
calcium was present in high concentrations, a mixture of the above mag-
nesium phosphate compounds and Ca(PO.)2 was precipitated. The composi-
tion of the precipitated solids was dependent on the amount of inert
impurities in the wastewater and on the relative concentrations of mag-
nesium and calcium present or added to the system. Where sufficient
cations such as calcium and magnesium are in the wastewater, addition
of these cations may not be required for the process.
A process for the precipitation of orthophosphate from digester super-
natant, based on the heat of decomposition of ammonium bicarbonate, was
tested on a bench and pilot scale (9). As the ammonium bicarbonate
decomposed, CO^ evolved from the system resulting in a pH increase.
Precipitation of calcium and magnesium phosphate was obtained without
need of chemical additives.
The above processes employing magnesium have the benefit of decreasing
both the phosphorus and nitrogen content of the wastewaters. At the
present time, the processes are in the developmental stage.
Alum and Iron Treatment - In case of lime coagulation, the principal
mechanism of phosphate removal is that of precipitation as insoluble
calcium phosphate salts. With iron salts and alum, adsorption upon
hydrated oxide floe particles plays a major role in addition to the
formation of an insoluble salt.
When alum is added to wastewater in the presence of alkalinity, the
following reaction occurs:
21
-------
A12(S04)3 + 6HCO~ -> 2A1(OH)3 + + 3S04 + 6C02 (4)
Aluminum hydroxide is a voluminous, gelatinous floe which enmeshes and
adsorbs colloidal particles thus providing clarification. In the pres-
ence of phosphates, another reaction also occurs:
A12(S04)3 + 2P04 + 2A1P04 i + 3S04
(5)
The above reactions compete for the aluminum ions. At pH values above
6.3, the phosphate removal mechanism is by incorporation in a complex
with aluminum or by adsorption on aluminum hydroxide floe. In addition
to these reactions, a substantial fraction of the phosphorus is in the
form of suspended and colloidal matter which can be removed by coagu-
lation rather than precipitation.
Because the above reactions take place when alum is used, the removal of
phosphorus is not in a direct stoichiometric relationship with alum. If
a stoichiometric relationship prevailed, 0.87 pounds of aluminum ion
would be required for each one pound of phosphorus removed.
Ferric and phosphate ions will react at pH values above 7 to form insol-
uble FePCL. The colloidal nature of the FePO. requires an excess of
ferric ion for the formation of a well-flocculating hydroxide precipitate.
The precipitate will include the FeP04 particles and other inorganics and
will act as an adsorbent for other phosphorus compounds. Efficient phos-
phorus removal requires the stoichiometric amount of iron, which is 1.8
mg Fe/£ per mg P/a, to be supplemented by additional iron for hydroxide
formation. Ferric chloride, ferric sulfate, and ferrous sulfate have
been chemicals used for phosphorus removal.
Recent investigations of the rate, mechanism, and stoichiometry of
phosphate precipitation with aluminum and ferric salts have been made
with pure phosphate solutions and municipal secondary effluents (10).
These studies showed that the reactions of the orthophosphate ion with
both Al (III) and Fe (III) are completed in less than 1 second. Lowering
of the reaction temperature from ambient to 5°C did not result in any
measureable change in the rate or extent of phosphate removal. In all
cases examined, phosphate removal from solution was accompanied by com-
plete precipitation of excess Al (III) and Fe (III) by hydrolysis
reactions. With an initial orthophosphate concentration of 12 mg P/a,
maximum phosphate removal was about 82% for both cations at a 2:1 ratio.
Coagulant Dose Determination - The quantity of coagulant to achieve spe-
cific phosphate removals depends upon characteristics of the wastewater
22
-------
I
such as pH, alkalinity, phosphate concentration and related factors that
affect the coagulant demand. Such factors vary from wastewater to waste-
water with the result that empirical relationships are used to estimate
the necessary chemical dose. Laboratory jar tests with a representative
sample of the specific wastewater enable an estimation of the required
quantity of chemical. As yet, general predictive relationships for
specific wastewaters have not been elucidated.
The results of the jar test experiments provide a point of departure for
proper chemical requirements which must be refined under actual operating
conditions. The jar test procedure represents controlled conditions such
as wastewater characteristics, degree of mixing, quiescent settling con-
ditions, and time of reaction, each of which may vary in a treatment
facility. Removal in practice can be poorer than estimated from jar test
experiments, especially since clarification characteristics of the resul-
tant suspensions can be different.
OBJECTIVES AND METHODS
The objective of this phase of the project was to evaluate the removal
of phosphorus from animal wastewaters by chemical precipitation. This
phase also accomplished the removal of color from these wastewaters. The
studies investigated the effect of phosphorus concentrations, pH control,
different coagulants, and wastewater source.
All experiments on poultry and dairy wastes were conducted in the agri-
cultural waste management laboratories, Cornell University. The wastes
used were untreated dairy manure and waste from laying hens from facili-
ties at the University. Because in practice wastes of this type would
be subject to removal of solids prior to any chemical addition, the
solutions used in these experiments were settled wastewaters. Tap water
dilutions of these settled animal wastes were used to obtain variations
in phosphorus concentration. The tap water used did not contain appre-
ciable amounts of phosphates.
Data from laboratory experiments on the removal of phosphates from duck
wastewater are included in this report to provide information on another
type of poultry waste and permit an extension and further interpretation
of the data obtained in this phase of the project. The duck wastewater
study was done at the Cornell University Duck Research Laboratory,
Eastport, Long Island, New York by Mr. Kenneth J. Johanson with the
guidance of Dr. Raymond C. Loehr (11). A variety of chemicals were
evaluated on about 88 random grab samples from 22 duck farms during
the 1970 duck production season.
The duck wastes were collected from duck farms that only produced ducks
and shipped the ducks to a central slaughtering operation, from a
central cooperative duck slaughtering and processing operation, and
from duck farms that produced and slaughtered their own ducks. The
wastes are described in this report respectively as: a) duck farm
23
-------
wastewater, b) duck processing wastewater, and c) duck farm wastes
possibly containing processing wastes. The latter designation was used
since it was not known whether the actual samples contained processing
wastes at the time of sampling. On all graphs illustrating the results
from the duck wastewater studies, each symbol represents a different
waste sample.
A laboratory jar test apparatus was used for all experiments noted in
this phase of the report. The following procedure was followed: a;
one liter of the actual waste or a suitable dilution was used, b)
appropriate chemical dosages were added directly to the liquid, c)
the liquid was mixed for one minute at a paddle speed of 100 rpm, d)
the liquid was then mixed at a paddle speed of 30 rpm for 20 minutes,
e) the mixing was stopped, the mixing paddles removed, and settling
allowed to take place for at least 20 minutes before samples were with-
drawn for appropriate analyses. In some experiments, the pH of the
liquid was adjusted to a specific level before the coagulants were added.
The tests were carried out on a Phipps-Bird mechanical stirrer using six,
one liter samples. The chemical concentrations used were chosen to
provide a range of phosphate removals from below 40% to over 90%.
The glassware used in the experiments routinely was hot acid washed. All
experiments were conducted at room temperature, 20-23°C.
Prior to addition of any coagulants, the wastewater was analysed for
alkalinity, pH, ortho and total phosphate, ammonia nitrogen, total
solids, and calcium and total hardness. These parameters were obtained
to characterize the waste samples and to develop predictive relationships
that may be utilized for phosphorus removal from various wastewaters.
The characteristics of the wastewaters are presented in Table I, Appendix.
Except where noted, analyses were conducted by techniques presented in
Standard Methods (12). The stannous chloride method for orthophosphate
analysis was used because of accuracy and minimum detectable concentra-
tion considerations. The majority of the samples were light to dark
brown in color which would interfere with the accuracy of colorimetric
determinations. The low minimum detectable concentrations possible with
the stannous chloride method permitted smaller samples to be taken and
diluted to appropriate analytical levels to avoid color interference.
Color interference did not occur in the phosphorus analyses utilized in
this report.
Total phosphorus analyses were done by a modification of the persulfate
oxidation method of Menzel and Corwin (13). For routine evaluation of
color removal, the change in percent transmittance at 425 my on a
Spectronic 20 colorimeter was used. The chemicals used were of analytical
grade.
As used in this report the term phosphorus refers to phosphorus as a con-
stituent of various organic and inorganic complexes and compounds, not
24
-------
to elemental phosphorus as a chemical substance. =The results of all
phosphate analyses are reported in terms of mg PO:/A. The terms phos-
phate, orthophosphate, and total phosphate refer to P04 as determined
by the above analyses. All graphs and figures should be interpreted in
this manner.
RESULTS
General - Three chemicals were used in these experiments: alum
A12(S04)3; lime - CaO; and ferric chloride - FeClg-e^O. Unless explicitly
noted in the report, use of the terms alum, lime, and ferric chloride
refers to the compounds in the above forms.
The removal patterns that occurred in these experiments with the addi-
tion of three coagulants, alum, lime and ferric chloride (Figure 3) are
typical of those obtained by other investigators and with other waste-
waters. The different wastewater samples required different coagulant
dosages to accomplish a specific phosphorus removal. In the wastewaters
examined in this study, the total phosphate removal curve either para-
lleled or converged on the orthophosphate removal curve. With lime as
the coagulant, the curves generally did not converge. As noted in the
Methods of Removal section, organic and polyphosphates are more difficult
to remove using lime, perhaps accounting for the necessity of high lime
dosages with these wastes.
When lime was used, color removal was poor. A brownish, turbid solution
resulted even at high lime dosages. The turbidity was due both to a
portion of the initial turbidity that remained and to lime that stayed
in the solution. Clarification of these supernatants occurred with
the addition of an iron salt. When lime and ferric chloride were used
together, a clear effluent generally resulted. This combination was
used in only a few experimental runs. The solids generated in all
experiments settled rapidly.
When alum was used and when ferric chloride was used on dilute poultry
wastewater, the color removal and phosphate remaining were similar.
For large scale units in which phosphorus is removed from colored solu-
tions, the color of the resultant effluent may be able to be used as a
routine operational and predictive tool of general process performance.
Correlation between color and the phosphate remaining would be needed
for each wastewater.
Concentration Effects - As expected, the greater the initial phosphate
concentration, the greater was the chemical dosage to obtain any
phosphate removal. The trends illustrated by poultry manure wastewater
are noted in Figure 4. Where two numbers are shown adjacent to specific
curves, the data from two runs having the initial orthophosphate con-
centration shown gave comparable results. Similar trends were observed
with dairy manure wastewater. The general pattern of removal is
25
-------
UJ
o
z
CO
5.5
0 50 100 150 200 250 300
RUN 55
UJ
O
a:
UJ
a.
tr.
o
o
o
Q.
CO 0
§ 0 200 400 600 800 1000 1200
0 100 200 300 400 500 600
CHEMICAL USED - mg/l
FIGURE 3
PHOSPHORUS REMOVAL PATTERNS FROM
POULTRY MANURE WASTEWATER
26
-------
1001
uj
CE 80-
UJ
a.
i
I 60
S
UJ
cr
UJ
< 40-
QL
-------
the inverse of the relationships noted in Figure 3.
Comparisons of the effect of initial phosphate concentration on the
quantity of chemical required are noted in Figure 5. These relation-
ships should be construed as indicative rather than precise. The data
for many runs were plotted and lines connecting similar percent removals
were estimated by eye. Not all the runs produced results that fit the
lines exactly. The patterns illustrate that for a given percent removal,
the chemical dosage varied directly with the orthophosphate concentration
in the untreated sample. Similar relationships were obtained for dairy
manure wastewater solutions.
For a given percent removal and initial orthophosphate concentration,
the required alum concentration was considerably less than that of lime.
For example, at an initial orthophosphate concentration of 300 mg/1 and
70% removal, the lime dosage was about 2200 mg/1 and the alum dosage
was about 550 mg/1 for the poultry wastewater. Comparable differences
in chemical dosage were apparent for the dairy manure wastewater.
The relative effectiveness of alum and lime were reversed in the duck
wastewater study (11). Alum requirements were greater than lime require-
ments for the duck wastewater.
A comparison of the relative chemical requirements for duck, poultry,
and dairy manure wastewaters is noted in Table 2. For the same removal
and initial phosphate concentration, the alum dosage was considerably
greater for poultry wastewater than for dairy manure wastewater at high
(greater than 100 mg/1) initial orthophosphate concentrations. At
lower initial orthophosphate concentrations, the alum dosages for dairy
and poultry wastewater were closer but were greater than for duck waste-
water at specific orthophosphate removals. At all initial orthophos-
phate concentrations, the lime dosages were greater than for alum for
poultry and dairy manure wastewater. The lime requirements were greatest
for poultry manure wastewaters and then for dairy and duck wastewater
in that order at comparable initial orthophosphate concentrations.
Each wastewater had its own chemical demand. The chemical demand of
poultry wastewater was larger than for dairy manure and duck waste-
water, and is due to constituents other than the phosphate concentration.
It is logical to expect that wastewaters from a specific agricultural
or industrial operation will have consistent chemical demands.
Relationships similar to those observed for orthophosphate removal also
were observed for total phosphate removal. Figure 6 presents the total
phosphate removals from dairy manure and poultry wastewaters using lime
and alum. The differences in chemical demand for the two coagulants
are apparent.
pH Effects - The pH of a solution affects chemical coagulation. Theriault
28
-------
f
300-
o
o
6
200
100
,40%
47 /50%
60%
60/°
70%
74 75 SI
/80%
POULTRY WASTEWATER
LIME
400 800 1200 1600 2000 2400
CHEMICAL DOSAGE - mg/l
500|
400
o
3004
Q.
i
tr
o
200-
100-
<07°V°^ >w .
22 49
"^99%
POULTRY WASTEWATER
ALUM
0 400 800 1200 1600 2000 2400
CHEMICAL DOSAGE mg/l
FIGURE 5
REMOVAL PATTERNS AS AFFECTED BY
INITIAL PHOSPHATE CONCENTRATION -
POULTRY WASTEWATER
29
-------
TABLE 2
ESTIMATE* OF CHEMICAL REQUIREMENTS FOR
PHOSPHATE REMOVAL FROM ANIMAL WASTEWATERS
Initial
ortho-
phosphate
cone, (mg/1)
Ortho-
phosphate
removal
CHEMICAL REQUIREMENTS - mg/1
WASTEWATER
Duck
Alum Lime
Poultry
Alum Lime
Dairy Manure
Alum Lime
50
70
100
200
300
60
80
90
60
80
90
60
80
90
60
80
90
60
80
90
70
100
110
180
110
130
35
60
65
70
85
95
100
225
300
125
260
350
175
320
425
290
500
780
425
680
1150
880
960
1400
2360
1820
100
140
145
120
160
180
130
210
240
190
380
460
260
560
630
260
670
1460
270
700
1540
290
650
1630
340
870
1980
390
1000
2340
if
from data obtained in this report and from data in Reference 11:
all alum data reported as A12(S04)3 and lime data as CaO.
30
-------
800-
400
UJ
o
<
CO
o
o
LIME
o - POULTRY
• - DAIRY
°
0
0
o
20
40
60
80
100
o
5
UJ
I
1600
1200-
800-
400-
ALUM
o - POULTRY
• - DAIRY
*
o o
o o
°
20 40 60 80 100
PERCENT TOTAL PHOSPHATE REMOVAL
FIGURE 6
TOTAL PHOSPHATE REMOVAL USING
LIME AND ALUM
31
-------
and Clark (14) demonstrated that optimum conditions for alum floe for-
mation were in the pH range of 5 to 5.5. Subsequent investigations have
clarified the effect of pH on the production of insoluble aluminum
complexes. Lea, et. al. (4) demonstrated that the optimum pH for the
removal of phosphates with alum from domestic sewage was in the range
of 7.1 to 7.7.
The quantity of alum required to remove phosphates from the concentrated
wastewaters depressed the pH of the solutions (Figure 3). The role of
pH in the removal of phosphates was investigated. In these studies,
the pH of the solutions was lowered to specific levels before the
varying alum concentrations were added in the jar test procedure. The
effect of pH control before alum addition can be observed in Figure 7.
This Figure suggests that with concentrated wastewaters, i.e., gen-
erally with initial orthophosphate concentrations greater than 120 mg/1,
the alum dosage decreased by 20-40% as the initial pH was decreased from
7.0 to 5.5. Variations caused by different wastewaters again can be
observed. Similar results were observed at other percent removals. The
differences were more apparent at the higher percent removals.
The greatest pH effect was apparent with the more concentrated waste-
waters. The possibility of reducing the required alum dosage by
decreasing the initial pH of a concentrated wastewater appears feasible
if one has a source of inexpensive acid available. The amount of acid
necessary to depress the pH to a given level in these experiments was
significant due to the high alkalinity of the concentrated wastewaters.
When the cost of the required acid and that of the decreased alum
dosage were combined, the combined cost of phosphorus removal by alum
at a controlled initial pH level was greater than that incurred if no
pH adjustment was made (Figure 8). Only chemical costs were utilized
in developing Figure 8 and the following chemical costs were used:
concentrated sulfuric acid (93.2% H2SO.) $3.25/100 Ib, commerical alum
(A12(S04)3-18H20 in 100 Ib. bags) $56/ton. Costs for small size chem-
ical purchases were used since it is unlikely that an animal production
operation would purchase the chemical in bulk quantities.
The combined cost increased as the initial pH of the solution was
lowered. The increase was most apparent with concentrated wastewaters.
At higher alum dosages, sludge formation and disposal can be a problem
which may be reduced by lowering the pH of the waste initially. Although
some cost is incurred by following such a practice, certain tradeoffs
exist between the amount of acid used and the quantity of chemical sludge
generated by the added alum. The initial lowering of the pH decreases
the alum requirement thereby reducing the quantity of sludge produced.
In situations where alum is the coagulant of choice and sludge disposal
is a significant problem, greater investigation of phosphorus removal
at lower initial pH levels may be warranted.
Predictive Unit Relationships * The chemicals used in this study remove
phosphates from a liquid by forming insoluble compounds that precipitate
32
-------
Q
UJ
Q
<
5
<
UJ
I
7.5-
7.0
6.5-
6.0
5.5
I
CL
5.0
INITIAL ORTHO-PHOSPHATE
/ CONCENTRATIONS
37,41
70,90
300,370
360
POULTRY WASTEWATER
200 400 600 800 1000
mg/l ALUM FOR 80% REMOVAL
1200
FIGURE 7
EFFECT OF pH CONTROL ON
ALUM DOSAGE
33
-------
§
o
LJ
CC
O o>
oo r>
%-
o
<
ii
L_
O
o
o
1.0-
.6-
.4-
.2-
0
POULTRY WASTEWATER
INITIAL
ORTHO-PHOSPHATE
CONCENTRATIONS
mg/1
300, 370
120,130
37 50
5.0 5.5 6.0 6.5 7.0
ADJUSTED PH OF SOLUTION
FIGURE 8
CHEMICAL COSTS FOR PHOSPHORUS
REMOVAL AT CONTROLLED pH VALUES
USING ALUM
34
-------
the phosphorus either as a co-precipitant or as adsorbed material on the
floe that is formed. Wastewater characteristics that may be related to
phosphate removal patterns include factors that affect the chemical
demand such as the alkalinity, ortho and total phosphate, total solids,
and calcium and total hardness of the untreated wastewater. With lime,
the resultant insoluble phosphate compound is related to the pH of the
solution after the lime is added. The effect of each of the above
parameters was investigated in the study. In order to permit comparison
of data from various runs, the appropriate data such as chemical dosages
are presented in non-dimensional ratios, i.e., mg/1 of chemical required
per mg/1 of the specific parameter.
a) Alkalinity - Equations 1 and 4 illustrate the relationship between
chemical dosage and the alkalinity of the wastewater. It might be
expected that this parameter would be useful in relating phosphate removal
to chemical dosage. Figure 9 portrays the chemical dosage required per
initial alkalinity of the untreated wastewater as related to percent
orthophosphate removal. A similar figure was obtained for total phos-
phate removal. While the differences between poultry and dairy manure
wastewaters are apparent with both chemicals, the patterns are clearer
when alum was used.
More definitive relationships were obtained with poultry and dairy
manure wastewaters than were obtained with wastes from duck production
operations (Figure 10). In Figure 10 each symbol represents a different
sample. The difference between duck farm wastewater, which contains
primarily duck manure, and duck processing wastewater, which contains
duck slaughtering and processing wastes, is quite apparent. Similar
results were obtained when alum was used on these wastes. Observable
differences are evident in the ratios of the chemical dosage per initial
alkalinity in Figures 9 and 10. While ratios of 1.0 or less were common
with poultry and dairy manure wastewater at all percent removals, the
ratios for duck farm wastewater ranged as high as 2 at high percent
removals and as high as 12 for duck processing wastewaters using lime.
No definitive relationships were observed when the chemical dosage per
initial alkalinity was compared to the orthophosphate remaining for
poultry and dairy manure wastewater samples although again differences
were evident for the two manure wastewaters. Very insensitive relation-
ships were obtained for duck wastewaters. The duck processing waste-
water again exhibited different relationships than did the duck farm
wastewaters.
The order of magnitude of the chemical dose per alkalinity when related
to orthophosphate concentration remaining was different for the different
wastewaters. For poultry and dairy manure wastewaters, the ratio of the
chemical dosage per initial alkalinity was generally less than 1.2 even
at low (less than 5 mg/1) concentrations of residual orthophosphate.
Ratios for duck farm wastewater generally were less than 1.0 but ranged
from 1-2 at low residual orthophosphate concentrations. Duck processing
wastewater ratios ranged up to 10 at low residual concentrations.
35
-------
1.0-
.8-
.6-
.4-
.2-
0
*
V)
o
Ul
o
.9 •
.8 •
.7-
.6-
.5-
.4 •
.3-
.2 •
.1 -
0 .
LIME
o - POULTRY
• - DAIRY
oo u
o
0 ° _ o
• o
0 °o
I
• o
• o
20
40
60
80
100
ALUM
•/• oo°X
DAIRY
POULTRY
20 40 60 80 100
PERCENT ORTHO-PHOSPHATE REMbVAL
FIGURE 9
PERCENT ORTHOPHOSPHATE REMOVAL RELATED
TO CHEMICAL POSE AND INITIAL ALKALINITY -
POULTRY AND DAIRY MANURE WASTEWATERS
36
-------
t 2.0J
DUCK FARM
WASTEWATER
FERRIC CHLORIDE
50
60
70 80 90 100
ui
2
CO
o
o
_l
<
o
5
UJ
o
8.0i
60J*
4.0-
2.0 J
PROCESSING
WASTEWATER
SAMPLES
o- DUCK FARM
WASTEWATER
50
60
70
LIME
80
90
100
PERCENT ORTHO-PHOSPHATE REMOVAL
FIGURE 10
PERCENT ORTHOPHOSPHATE REMOVAL RELATED
TO CHEMICAL DOSE AND INITIAL ALKALINITY -
DUCK WASTEWATER
37
-------
These relationships (Figures 9-10) illustrate that the alum dosage per
initial alkalinity ratio may be useful for poultry and dairy manure
wastewater when compared to percent phosphate removal to predict neces-
sary alum dosages. It does not appear that the chemical dosage per
initial alkalinity is a useful predictive relationship for other waste-
waters or other chemicals. Better relationships are necessary for
design and operational use.
b) Phosphate - The chemical dosage was related to the initial phosphate
concentration (Figure 5). The results of comparing the chemical dosage
per initial phosphate concentration produced patterns similar to those
of chemical dosage per initial alkalinity. Representative results are
presented in Figure 11. Al/P ratios for dairy manure wastewaters were
in the range of less than 3 up to 90% orthophosphate removal but gen-
erally ranged from 2-7 for poultry wastewaters at percent removals
above 50%. These ratios are higher than those of 1.5-3 generally
observed with municipal wastewaters.
Although specific patterns were obvious when the results from one jar
test run were plotted as a function of the initial phosphate concentra-
tion, the patterns were variable when results from many runs were com-
puted. The development of predictive relationships based upon the ratio
of chemical dose per initial phosphate concentration for these waste-
waters does not appear promising; however, Figure 11 does provide an
estimate of the ranges of such ratios that occur with these wastewaters.
A more useful predictive relationship was obtained when the chemical
dosage per remaining phosphate concentration was plotted versus the
percent phosphate removed. Typical relationships are presented in
Figures 12 and 13. With poultry and dairy manure wastewater, the data
points using alum fit the relationships better than the data obtained
with lime. The relationship with total phosphate using lime also demon-
strated greater variability than with alum.
Similar relationships were obtained with duck wastewater (11) for both
alum, lime, and ferric chloride. Figure 13 typified the results using
alum with the duck wastewater. In each of the Figures, the relationships
appear asymptotic near 100% removal demonstrating again the difficulty
of obtaining high percent removals.
In each of the relationships of chemical dosage per remaining phosphate
concentration versus percent phosphate removal (Figures 12 and 13), data
for a number of laboratory runs were plotted. The variation among the
runs can be observed in the duck wastewater results in Figure 13. The
fact that these relationships are reasonably sensitive for data from a
number of different wastewaters, for different concentrations of phos-
phates, and for three different chemicals suggest that these relation-
ships may be fundamental in nature and useful for practical application.
These relationships should be explored more fully with other wastewaters
38
-------
8
7
^V 6
15 -
2»
Q
5j
{£ 10-
5
o
o
p
T
(O
O
X
i °J
\
. < PROCESSING g
WASTEWATER p 5 .
a:
t_
/ Z 4.
/* "
LIME , ^ . 0 ,
^^ 4 o °'
DUCK BWM ^ . u
„ WASTEWATER^. 4 " . t 2-
^*-— "^" • *• I
_-- "^^ i + « * i
__^-i-— t ." » f g |.
& -•. jt ^& * °-
• & * »* J__-— - o 0
50 6O 70 BO 90 KM o
o ^
OJ <
40 < 30-1
t
z
^
o 20
O
J
I »•
Ill
5
0
t 6 -
+ 2
FERRIC CHLORIDE ^
m 5 •
O
jf^ O A
DUCK FARM ,x Q * '
POSSIBLE .,+ WASTEWATER / « _,
PROCESSING — \ ^' o 3 •
WASTEWATER 3" I
— c> HI
_^^^^ * * 52-
-Ota •*" T 1
' "*" 4
* — * -» ^ ^ — 0
o
• -POULTRY * m
o- DAIRY .
o
o
o
•
0
Q
* »
00 0. •
® •
• *. °* LIME
•
20 40 60 SO IOO
0 o
o-POULTRY °
• -DAIRY o o
0
0 0 ° ° 0
< 0 .
00.'
00 ° •
0 0 0
0 • *
0 0 0
o ° 9
0 •
•o o . *
0
a
o° * ° * ALUM
o
50 60 70 8O 9O IOO
PERCENT ORTHO-PHOSPHATE REMOVAL
20 40 60 80 IOO
PERCENT ORTHO-PHOSPHATE REMOVAL
FIGURE 11
PERCENT ORTHOPHOSPHATE REMOVAL RELATED TO
CHEMICAL DOSE AND INITIAL ORTHOPHOSPHATE
-------
0 20 40 60 80 100
PERCENT ORTHO-PHOSPHATE REMOVAL
100
20 40 60 80 100
PERCENT TOTAL PHOSPHATE REMOVAL
FIGURE 12
PERCENT PHOSPHATE REMOVAL RELATED TO
CHEMICAL DOSE AND REMAINING PHOSPHATE
POULTRY AND DAIRY MANURE WASTEWATERS
40
-------
80 40 60 80 100
PERCENT ORTHO-PHOSPHATE REMOVAL
I"
i
JK
s
i
HUN SYMBOL
33 »
|
f X
DUCK WASTEWATER
SO 60 70 BO 90 100
PERCENT 0-P04 REMOVAL
FIGURE 13
PERCENT PHOSPHATE REMOVAL RELATED TO
CHEMICAL DOSE AND REMAINING PHOSPHATE -
POULTRY, DAIRY, AND DUCK WASTEWATERS
41
-------
since if they are valid, they offer a simple approach for the design
and operation of chemical precipitation of phosphates from wastewaters.
With percent removal requirements established and influent phosphate
concentrations known, a curve such as shown in Figures 12 and 13 would
establish the chemical dosage to meet the expected removals. Because
influent phosphate concentrations and chemical demand relationships
can be expected to differ for various wastewaters, a specific relation-
ship such as noted in the Figures 12 and 13 may have to be developed
from laboratory and operating data at each location.
These relationships appear more precise than others explored in this
phase of the project and would be preferable as both design and operating
parameters. Orthophosphate analysis is not difficult and field ortho-
phosphate analysis kits are available for use by operating personnel.
Routine phosphate analyses by plant personnel in conjunction with rela-
tionships such as in Figures 12 and 13 could help attain desired effluent
concentrations or percent removals.
c) Total Solids - The chemical demand of a wastewater is due to many
factors and it was possible that the chemical dose may be related to the
total solids concentration, a gross parameter which includes many
chemical components. A plot of the chemical dosage per initial solids
concentration was made for each of the wastewaters and for the different
chemicals used. Very insensitive relationships were obtained indicating
that total solids was not a useful parameter for predicting phosphate
removal s.
An example of the type of relationship obtained is shown in Figure 14
for duck wastewater. Each symbol represents different jar test runs.
Although a pattern may be observable for a specific run, no predictive
pattern resulted when all the data were compared.
d) Hardness - Calcium and magnesium ions, the principal substances
causing hardness, are important in phosphorus precipitation reactions
(Equations 1-4). It might be expected that chis parameter would be
useful in establishing predictive relationships for phosphate removal.
Both calcium and total hardness were used to investigate this possibility.
Results typical of orthophosphate removal are noted in Figure 15. Similar
results were obtained for alum and for total phosphate removal. These
relationships were obtained only for poultry and dairy manure wastewater.
The relationships for hardness are similar to that obtained for alkalinity
(Figure 9) and orthophosphate (Figure 11) but appear more precise than
the relationships for the other parameters. These results indicate that
either calcium or total hardness may be suitable predictive measures of
the chemical dosage to obtain specific phosphate removals. The validity
of the hardness relationships should be explored more fully since it
appears that they may be useful for design and operation. Hardness
analyses are not difficult and also can be accomplished by laboratory
42
-------
.41
FERRIC CHLORIDE
.2-
* * -*-
o 50
60
70
80
90
100
to
Q
.8
ALUM
O
O
.4-
£ .2-1
o
-s-
A "
*
.-*
-*-
-A- *
I *
-A-
50 60 70 80 90
PERCENT ORTHO-PHOSPHATE REMOVAL-
FIGURE 14
PERCENT ORTHOPHOSPHATE REMOVAL RELATED TO
CHEMICAL DOSE AND INITIAL TOTAL SOLIDS -
DUCK WASTEWATER
43
100
-------
141
UJ
CO
O
o
g
UJ
o
(T
<
X
S
6
O
£ 2
LIME
o
o
o
o
0 20 40 60 80
PERCENT ORTHO-PHOSPHATE REMOVAL
FIGURE 15
PERCENT ORTHOPHOSPHATE REMOVAL RELATED TO
CHEMICAL DOSE AND INITIAL HARDNESS -
POULTRY AND DAIRY MANURE WASTF.WATER
100
44
-------
and field methods to assure desired removals when combined with relation-
ships such as in Figure 15.
e) pH Effect - There is a general relationship between the lime dosage,
the alkalinity of a wastewater, and the pH of the solution after the
chemical addition. Because the insolubility of the calcium phosphorus
compounds is a function of the pH of the wastewater, relationships to
predict the pH are useful. The pH after lime addition is controlled by
the buffering capacity of the wastewater through neutralization of the
hydroxyl ions. The lime dose required to overcome the buffer capacity
should increase with increasing alkalinity of the wastewater since
wastewater alkalinities usually measure bicarbonate-carbonate, acid
phosphate-phosphate and ammonium-ammonia buffer capacities.
A different range of lime to alkalinity ratios to obtain a specific pH
may exist for different wastewaters. The data of Buzzell and Sawyer (7)
indicated that the lime requirement, as CaO, to raise the pH of municipal
wastewaters to 11 ranged from 1.0 to 1.9 times that of the alkalinity in
the wastewaters. In their studies, high phosphorus removals were accomplished
The poultry and dairy manure wastewater studies showed that the lime to
alkalinity ratio to raise the pH of these wastewaters to 11 was approxi-
mately 1.0-1.1 (Figure 16). Both poultry and dairy wastewaters exhibited
the same pattern. For duck wastewaters the ratio varied from about 1.5
to 3. Although different runs with the duck wastewater exhibited specific
patterns, pooling all the data produced an overall pattern that was less
distinct. Alkalinity obviously is not the only parameter to exert a lime
demand. Other parameters such as hardness, phosphate, and colloidal
organic matter also exert a coagulant demand.
The relationship between the pH of a solution and the removal of phos-
phorus with lime has been demonstrated by a number of investigations.
Figure 17 illustrates typical relationships for the wastewaters included
in this study. Figure 17A, B, and C represents three different poultry
wastewaters while Figure 17D represents a dairy wastewater. The pH
values noted in the Figures represent the pH of the wastewater after the
lime was added but before the mixing in the jar test procedure was started.
The pH after mixing decreased by 0.1 to 0.5 pH units in many of the runs.
The larger decreases occurred in the wastewaters that had the higher
initial pH levels. Because of the small volume used, one liter, pH con-
trol during the jar test procedure was not feasible.
A number of observations can be made from Figure 17. The effect of the
wastewater concentration again can be seen. In general, there were lower
percent removals in the wastewaters that had lower initial orthophosphate
concentrations. For a given wastewater, the percent removals for solu-
tions having initial orthophosphate concentrations above 100 mg/1 were
reasonably consistent irrespective of initial concentration.
With the exception of one dilute wastewater (Figure 17B), high removals,
45
-------
1.0'
.9-
i -8
I-7-
_» .6-
I-
I .5-
^
(9 A-
iO •
UJ
1 .2-
.1 •
0.
•/
v o
/o
0 /
7
o /
/
°°/
/o
/•
• -DAIRY
o- POULTRY
o° oo
7 ' 8 ' 9 ' 10 II 12
pH AFTER LIME ADDITION
4
3
>-
LIME DOSAGE
IAL ALKAUINI
ro
— t-
I
C
p
DUCK WASTEWATER
/ V
/».» /
/*. 0 /
X ! °* -/
/.+*./
>•• c • •» -e-
•^ *^ " /
^/^ DO • » /
0--^ » /
-^^ *b ** Jr^
^^ 0 * s^
•a- ^^^ «. * -
^^ g. •* • ^-^ —
8 9 10
pH AFTER LIME ADDITION
12
FIGURE 16
LIME DOSE RELATED TO INITIAL ALKALINITY
AND pH AFTER LIME ADDITION
46
-------
lOOl
o
Q.
CO
O
50
637
ao
9.0
10.0 11.0
1001
50
91
41
'20
(B)
8.O 9.0
K>.0 11.0
§ 100
o
-------
above 90%, rarely were obtained even at high pH levels. A part of the
problem is due to the fact that actual removals were greatly dependent
on the clarification characteristics of the calcium precipitate sus-
pension. While considerable phosphorus removal was achieved by high
lime dosages, the treated effluent from these studies contained con-
centrations of phosphate that could cause difficulties in receiving
waters. It is possible that removal of the remaining fine solids in
the supernatant could have been removed by filtration or other more
efficient solids separation methods.
If the criteria for phosphorus removal are based on percent removal, the
data in Figure 17 illustrate that, for concentrated wastewaters, con-
siderable removal using lime and solids separation by sedimentation,
60-80%can be obtained with pH control in the range of 9.0 to 9.5.
The engineering implications are obvious in both the quality of lime
needed and with the problem of ultimate sludge disposal. The flatness
of the shape of the curves in Figure 17 indicated that for these con-
centrated wastewaters, there was negligible advantage in adjusting the
pH to a level higher than 9.0 to 9.5.
f) Summary of Predictive Relationships - A number of relationships were
investigated to determine those that could be useful in predicting chem-
ical dosages and residual phosphate concentrations or percent phosphate
removals. A comparison of the predictive relationships that were
examined indicated that the most sensitive parameters were:
i) -chemical dosage per remaining total or orthophosphate concen-
tration versus percent total or orthophosphate removal for alum, lime
and ferric chloride for poultry, dairy manure, and duck farm wastewaters
ii) -chemical dosage per initial calcium and total hardness versus
percent total and orthophosphate removal for poultry and dairy manure
wastewaters
iii) -lime dosage per initial alkalinity versus the pH after lime
addition for poultry and dairy manure wastewater
iv) -chemical dosage per initial alkalinity versus percent orthophos-
phate removal for alum with dairy manure and poultry wastewater
v) -chemical dosage per initial total phosphate concentration versus
total phosphate remaining for alum with dairy manure and poultry waste-
water.
Of all the relationships, the chemical dosage per phosphate remaining
ratios and the chemical dosage per initial hardness ratios appeared
most sensitive and narrow.
Other Parameters - Chemical precipitation of wastewater will remove con-
taminants other than phosphates. It was expected that the organic
48
-------
fraction will decrease and would be observed as a decrease in COD. In
these studies the COD removals appeared unrelated to either the chemical
dosage or the percent phosphate removal. Removals ranged from 20 to 70%.
Some ammonia nitrogen removal occurred but was small ranging from 0-20%.
Large ammonia nitrogen removals are not expected since ammonia is a
soluble constituent not readily removed by chemical precipitation.
Some organic removals will occur in any chemical precipitation system
and will provide a secondary benefit. When used on the effluent from
a secondary treatment system, the resulting effluent not only will con-
tain less phosphate but also less organic matter and should have a
smaller chlorine demand.
Solids Production - An evaluation of the appropriate chemical is incom-
plete if based solely on required chemical concentration. Of equal
importance is the quantity of solids generated for ultimate disposal.
During the investigations using the duck farm wastewater, an estimate
of solids production was made by measuring the suspended solids con-
centration of the mixed contents of the jar test experiments. The
suspended solids increase per quantity of chemical (solids increase
ratio) was used to estimate the solids production.
No definitive relationships were obvious. The average mg/1 solids
increase per mg/1 chemical added ranged from 0.1-1.9 for alum as
A12(S04)3-18H20, from 0.4-2.5 for lime as CaO, and from 0.1-2.5 for
ferric chloride as FeCl3-6H20. The majority of the solids increase
ratios were in the range of 0.6-1.0. The results provide a general
estimation of the additional sludge that will have to be handled and
disposed of when chemicals are used for phosphate removal.
The sludge production was compared (11) to the characteristics of the
wastewaters in an effort to obtain other predictive relationships. The
spread of data was great and again few definitive relationships were
obvious. The best relationship occurred between the solids increase
ratio and the initial alkalinity of the wastewater. The relationship
with alum was broad but reasonably definite. A less definite relation-
ship was observed with lime and no relationship was observed with ferric
chloride.
More information on the amount of sludge that will be produced will have
to result from larger scale studies.
Comparison of Chemicals - It is difficult to directly compare the three
chemicals used in these investigations since they have been used with
wastewaters of different characteristics and chemical demand. The
factors that govern chemical choice in a treatment facility are waste-
water characteristics, effluent discharge requirements, plant size, chem-
ical costs, sludge handling and disposal facilities, and other processes
used at the facility.
49
-------
A few general comparisons were made in this study. The poultry wastes
usually contained large concentrations of sulfides due to the anaerobic
conditions in the holding pits at the poultry house and in the sample
container before use. A black color was generated when ferric chloride
was added due to the formation of iron sulfide. High iron dosages, at
times greater than 200 mg Fe/1, clarified the solutions. The use of
alum and lime did not create these or other color situations in the
waste samples.
The chemicals in this study can be compared only on the costs of chem-
ical cost and sludge production. The three chemicals were used with
specific poultry and dairy manure wastewaters to provide a reasonable
comparison of chemical costs. Percent removal and chemical dosage
data were obtained from results of specific runs such as those noted
in Figure 3. Results from this comparison are presented in Table 3.
Lime was unable to obtain phosphate removals higher than 80% with wastes
A and C. Ferric chloride and lime with ferric chloride also were com-
pared directly in one wastewater, waste B. The comparison points out
the need for evaluation of costs and chemicals on a specific wastewater
rather than making sweeping generalizations since no distinct patterns
emerged. The chemical costs of ferric chloride in this comparison were
greater than those of alum and lime. Where the chemical costs of lime
and alum were approximately the same (waste A and C), alum would be the
chemical of choice since the amount of alum needed was less than that of
the lime and less solids would be generated for ultimate disposal.
Where the required lime dosages were less than that of other chemicals,
lime is the obvious chemical of choice since both chemical cost and
sludge production would be less. The information in Table 3 should be
used to compare the various chemicals and not to obtain typical cost
estimates of phosphate removal from animal wastewaters.
Cost Relationships - The characteristics of the poultry and dairy manure
wastewaters were varied over a large range to study concentration and
chemical dosage relationships. Hence, the characteristics of the samples
were not those considered "typical" of wastewaters from actual liquid
poultry and dairy manure wastewaters or treatment systems. Definitive
cost relationships are difficult to obtain with such an approach. How-
ever, estimates of cost relationships can be obtained by: a) using the
predictive relationships developed in this study with average waste
parameters, and b) using the predictive relationships and results
obtained in the duck wastewater study.
Average data on the total dry solids and P205 in the wastes from dairy
and poultry indicate that these values are about 8# Total Solids (TS)/
animal/day and 0.12# P205/animal/day for dairy cattle and 0.066# TS/
bird/day and 0.0026# P205/bird/day for poultry (15). A range in these
parameters has been reported for most dairy and laying hens and hence
these values are useful only as an average to estimate the costs of
phosphate removal.
50
-------
Waste
Poultry
(A)
Initial
Ortho-
Phosphate
Cone, (mg/1)
300-330
Poultry
(B)
75-115
Dairy
(C)
110-120
TABLE 3
COMPARISON OF CHEMICALS
Ortho-
Phosphate
Removal
70
80
Alum
Lime
Ferric
Chloride
Lime and
Ferric Chloride
(CaO) (Fe)
-Chemical Dosage* - mg/1-
800 4250
1080 4400
-Chemical Costs** - ($/10° gal.)-
70 390 390
80 530 400
70
80
90
-Chemical Dosage - mg/1-
420 250
490 350
550 1050
70
80
90
70
80
-Chemical Costs - ($/10c
210 23
240 32
270 96
gal.)-
70
80
Chemical Dosage - mg/1
190 650
225 800
-Chemical Costs - ($/106 gal.)-
90 60
110 73
150
180
210
160
160
160
80
90
150
310
370
430
180
200
320
*Alum as A12(SO.),, lime as CaO, ferric chloride as Fe '. When lime and ferric chloride were used in com-
bination, Time was added to raise the pH of the solution to 8.5 before the ferric chloride was added.
**These values were based on the following chemical costs:
as A12(S04)3-18H20, and ferric chloride at $170/ton.
lime (90% CaO) - $20/ton, aluminum sulfate $60/ton
-------
There are no typical amounts of water that dairy or poultry operators
add to their waste to handle them in a liquid manner. Enough water is
added to make the resultant slurry pumpable by available equipment.^
Slurries containing 10% solids or more have been pumped at these animal
production operations. For the purposes of these cost estimates, three
slurries were utilized, i.e., approximately 1, 5, and 10% total solids.
The 5 and 10% solid slurries represent conditions such as those that
might exist in actual animal production operations and the 1% slurry
may represent a dilution similar to a very strong industrial waste.
Table 4 outlines the characteristics of the three slurries for each
waste that were used in this estimation. It was assumed that the P205
concentrations were the equivalent of the total phosphate concentration
in the resultant dilutions. The cost estimations were made in terms of
total phosphate as P04 by modifying the P205 concentrations appropriately.
Table 5 and 6 indicate the chemical concentrations, amounts, and unit
costs associated with specific degrees of total phosphate removal. It
was assumed that the predictive relationship in Figure 12 and a similar
one for lime were applicable with the dilutions that were assumed. The
Figure used for lime was developed from data obtained in this study.
Total phosphate removals greater than 90% from poultry wastewater with
lime and greater than 95% for both wastewaters with either alum or lime
were not able to be estimated from relationships such as in Figure 12.
The lack of data points in the above 90-95% removal range caused diffi-
culty in estimating chemical dosage ratios in these ranges. The chem-
ical costs used for the estimate were those noted in Table 3: alum -
$60/ton as A12(S04)3«18H20 ($115/ton as A12(S04)3) and lime - $20/ton.
The cost estimates noted in Tables 5 and 6 should not be viewed as
realistic. It certainly would be folly to chemically treat waste solu-
tions containing 5 and 10% or even 1% solids. There are more applicable
methods of phosphate control for these slurries, namely controlled land
disposal. The estimates were made only to provide some measure of the
costs that could be involved when using chemical precipitation for
phosphate control from such wastes. Chemical precipitation is more
applicable to dilute wastes.
Tables 5 and 6 are useful in estimating the range of chemical costs and
in demonstrating the use of predictive relationships such as in Figures
12 and 13. The chemical costs per 100 dairy cattle and 1000 birds are
reasonably low, ranging from $0.10 to $1.10 per day for 100 dairy cattle
and $0.05 to $0.40 per day for 1000 birds, depending upon the chemical
used. Most dairy herds are of 100 head or less. However, most large
poultry operations contain upwards of 30,000 birds per operation and
some complexes contain up to 300,000 to 1,000,000 birds. Sizable costs
V!'- U5J° $12?/day for chemical costs for a 300,000 bird operation to
obtain 90/0 total phosphate removal would be involved.
The chemical costs per 1000 gallon decrease due to dilution; however, the
52
-------
TABLE 4
CHARACTERISTICS OF ANIMAL WASTE SLURRIES
ASSUMED FOR CHEMICAL COST ESTIMATES
IN THE PHOSPHORUS REMOVAL STUDY
POULTRY
MANURE
DAIRY CATTLE
MANURE
INITIAL WASTE CHARACTERISTICS - #/ANIMAL/DAY
Total Dry Solids
P2°5
(P04)
SLURRY CONCENTRATION
(% TS)
WATER REQUIRED FOR
DILUTION (gal.)
TOTAL PHOSPHATE*
(mg/1)
10
80
iOO
0.066
0.0026
(0.0017)
5 1
160 800
1300 260
10
8
0.12
(0.08)
5 1
1000 2000 10000
960 480
96
*Resulting from solids from 1000 birds or 100 head of cattle diluted by
the volumes noted.
53
-------
en
TABLE 5
COST PROJECTIONS FOR PHOSPHATE REMOVAL FROM POULTRY MANURE WASTEWATER
Waste Solution
TS)
10%
1%
Total Phosphate
Removal (%)
Total Phosphate
Remaining (mg/1)
LIME
Dosage/T-P04
Remaining
mg/1 required
# required/day
CHEMICAL COST
-$/day
-/WOO birds/day
-$106 gal.
-tf/1000 gal.
ALUM
Dosage/T-PO,
Remaining
mg/1 required
# required/day
CHEMICAL COST
-$/day
-$/1000 birds/day
-106 gal.
-(t/1000 gal .
50
1300
6
7800
5.2
.052
.052
650
65
3.5
4550
3.0
.17
.17
2100
210
70 90 95 50 70 90 95 50 70 90 95
780 260 130 650 390 130 65 130 78 26 13
20 100 - 6 20 100 6 20 100
15,600 26,000 - 3900 7800 13,000 - 780 1560 2600
10.4 17.3 - 5.2 10.4 17.3 - 5.2 10.4 17.3
.104 .173 - .052 .104 .173 - .052 .104 .173
.104 .173 - .052 .104 .173 - .052 .104 .173
1300 2200 - 325 650 1080 - 65 130 220
130 220 - 32 65 108 - 6.5 13.0 22.0
8.4 40 68 3.5 8.4 40 68 3.5 8.4 40 68
6550 10,400 8840 2275 3275 5200 4420 455 655 1040 885
4.4 6.9 5.9 3.0 4.4 6.9 5.9 3.0 4.4 6.9 5.9
.25 .40 .34 .17 .25 .40 .34 .17 .25 .40 .34
.25 .40^ .34 .17 .25 .40 .34 .17 .25 .40 .34
3100 5000 4250 1100 1560 2500 2100 210 310 500 425
310 500 425 110 156 250 210 21 31 50 42.5
-------
en
en
TABLE 6
COST PROJECTIONS FOR PHOSPHATE REMOVAL FROM DAIRY CATTLE MANURE WASTEWATER
Waste Solution
(% TS)
10%
5%
1%
Total Phosphate
Removal (%)
Total Phosphate
Remaining (mg/1)
LIME
Dosage/T-P04
Remaining
mg/1 required
# required/day
CHEMICAL COST
-$/day
-$/100 head/day
-$106 gal.
-tf/1000 gal.
ALUM
Dosage/T-PO.
Remaining
mg/1 required
# required/day
CHEMICAL
-$/day
-$/100 head/day
-$106 gal.
-(t/1000 gal.
50
480
2.5
1200
10.0
.10
.10
100
10
1.3
610
5.2
.30
.30
300
30
70
288
5.8
1660
13.9
.14
.14
140
14
4
1150
9.6
.55
.55
550
55
90
96
34
3260
27.2
.27
.27
272
27.2
24
2300
19.2
1.10
1.10
1100
110
95
48
70
3360
28.0
.28
.28
280
28.0
42
2020
16.8
.96
.96
960
96
50
240
2.5
600
10.0
.10
.10
50
5.0
1.3
305
5.2
.30
.30
150
15
70 90
144 48
5.8 34
830 1630
13.9 27.2
.14 .27
.14 .27
70 138
7.0 13.8
4 24
575 1150
9.6 19.2
.55 1.10
.55 1.10
275 550
27 55
95
24
70
1680
28.0
.28
.28
140
14.0
42
1000
16.8
.96
.96
480
48
50
48
2.5
120
10.0
.10
.10
10
1.0
1.3
61
9.2
.30
.30
30
3.0
70
29
5.8
166
13.9
.14
.14
14
1.4
4
115
9.6
.55
.55
55
5.5
90
10
34
326
27.2
.27
.27
27
2.7
24
230
19.2
1.10
1.10
no
11.0
95
5
70
336
28.0
.28
.28
28
2.8
42
202
16.8
.96
.96
96
9.6
-------
size of the needed waste treatment facility increases in direct relation
to the quantity of water to be treated offsetting the reduction in
chemical costs. Even at the lowest dilution used in the estimate, 1%
TS, the chemical costs are frequently an order of magnitude or more
larger than chemical costs quoted for phosphate removal from municipal
wastewaters.
Tables 5 and 6 also illustrate the use of relationships such as shown in
Figures 12 and 13. The curves representing these relationships should ,
be drawn with care to best represent the data. The curves that were used
provided ratios at the 95% removal level that indicated that lower
chemical dosages were required at 95% than at 90% removal. Obviously
this is incorrect and results from lack of adequate data points at the
higher percent removals. Care should be taken when using such predic-
tive relationships at high percent removals.
The least cost chemical for the poultry and dairy manure wastewaters was
lime even though the required alum concentrations were less than those
for lime. The cost of alum was significantly higher than that for lime.
Sludge disposal requirements for lime are likely to be greater than for
alum due to the greater quantities of lime that were needed.
More realistic cost relationships for the removal of phosphates from
animal wastewaters were possible using the data from the duck waste-
water study (11). The characteristics of the duck wastewater are less
variable and the samples from this study represented wastes emitted
from full scale field facilities. The average duck wastewater charac-
teristics used in this analysis were: 34 mg/1 of orthophosphate, a flow
of 178,000 gallons per day and an average water use of 15 gal./duck/day.
The results of the 1970 Laboratory Study on Duck Wastewater (11), using
costs of chemicals obtainable on Long Island, indicated that lime was
the chemical of choice in terms of both costs and chemical requirements.
Over the orthophosphate removal range of 50-90%, the chemical costs of
lime were estimated to be from 0.7-4.3 cents/1000 gallons of waste/day.
The chemical cost range for alum was 2.2-4.3 and for ferric chloride
was 9.1-25.5 cents/1000 gallons of waste/day.
The alum and lime costs are similar to slightly higher than the chemical
costs observed with municipal wastewater, i.e., about 1.5-2 cents/1000
gal. for lime and 3-4 cents/1000 gal. for alum for 90% phosphate
removals and a phosphate residual of less than 0.5 mg/1. Chemical costs
for ferric chloride in the duck wastewater study are considerably higher
than those reported for municipal wastewaters. It should be noted that
although chemical costs and orthophosphate percent removals may be
similar between duck wastewaters and municipal wastewaters, the "average"
duck wastewater assumed in these examples would have a residual ortho-
phosphate concentration of about 3.5 mg/1 at 90% removal.
The duck wastewater costs represent wastes with average waste quantity
56
-------
and quality as determined from the 1970 study (11). Actual costs can
be expected to vary at individual duck farms. However, the costs do
provide an estimate of the more feasible chemicals and the chemical
costs associated with phosphate removal. The least cost chemical at
all percent removals for the duck wastewater was lime, followed by
alum and ferric chloride in that order- Although more realistic than
the previous costs for poultry and dairy manure wastes, the above costs
for phosphate removal from duck wastewater should be viewed as estimates
rather than precise costs. The unit costs represent only potential
chemical costs and do not include power costs for mixing, or the costs
of additional units and equipment which will be necessary for chemical
addition and phosphate removal, or the costs of sludge handling and
disposal.
In a brief evaluation of the combined use of ferric chloride and lime
with duck wastewaters, it was observed that the percent phosphate
removals obtained by the combination of chemicals was similar to that
obtained in other experiments when only the same lime dosages were used
to remove the phosphates. This brief evaluation suggested that ferric
chloride may not be effective in combination with lime to remove phos-
phates from animal wastewaters.
The results from the chemical and cost comparison study indicated that
lime appeared to be the most economic chemical to be considered for
phosphate removal from poultry, dairy and duck wastewaters. An eval-
uation of alum may be warranted with poultry and dairy wastewaters
because of solids production and disposal considerations since lower
chemical concentrations were required with alum than with lime.
SIGNIFICANCE OF THE RESEARCH
Three chemicals, alum, lime, and ferric chloride, were used to evaluate
the removal of phosphates from animal wastewaters. Laboratory jar tests
were conducted on settled poultry and dairy manure wastewaters to
determine the appropriate chemical, the effect of these chemicals on
wastewaters of varying characteristics, and predictive relationships
that could be used for design and operation purposes. Data from a
similar study on duck farm wastewaters were included in the report to
compare and extend the data obtained with the poultry and dairy manure
wastewaters.
One of the purposes of the study was to evaluate color removal from
these wastewaters. When lime was used, color removal was poor. Residual
turbidity and color was due to portions of the initial turbidity and
color and to lime that stayed in solution. The residual color and
turbidity could be removed by addition of an iron salt. When alum was
used with all wastewaters and when high concentrations of ferric chloride
were used with dilute poultry wastewaters, color removal was good. When
ferric chloride was used with wastes having a high sulfide content, such
as poultry wastewaters, iron sulfide was produced resulting in a black
57
-------
solution darker than the original color. High iron dosages were neces-
sary to obtain a clear solution.
The concentration of alum necessary to obtain a specific percent phos-
phate removal from poultry and dairy manure wastewater was less than
that required for lime. However, with duck wastewater, required lime
dosages were less than alum.
Each wastewater had its own chemical demand relationship. The chemical
demand appeared to be in proportion to parameters such as initial
phosphate, alkalinity, or hardness concentrations in the wastewaters.
Studies on lowering the pH of wastewaters before phosphorus removal
indicated that there is value in decreasing the initial pH when using
alum, especially when treating concentrated wastewaters. Smaller con-
centrations of alum were needed at lower pH levels. With concentrated
wastewaters the alum dosage decreased by 20-40% when the initial pH was
decreased from 7.0-5.5. The practical effect is small due to the quan-
tity of acid that is needed unless an inexpensive supply of acid is
available. If sludge disposal is a significant problem, phosphorus
removal at depressed pH levels may be warranted since decreased alum
dosage results in less sludge for ultimate disposal.
Phosphorus removal with lime is due to the formation of calcium phos-
phates, the insolubility of which is a function of the pH of the solution.
The ratio of the lime required per unit of initial alkalinity to obtain
a specific pH has been used as a predictive parameter. Ratios of from
1.0-1.9 have been observed to raise the pH of municipal wastewaters to
11. Similar ratios for poultry and dairy manure wastewaters to raise
the pH of these wastewaters to 11 were about 1.0-1.1. The ratio varied
from about 1.5-3.0 for duck wastewater.
When the relationship between the pH of a waste and the removal of
phosphates with lime was investigated, high phosphate removals were not
obtained at high pH levels. For a given wastewater, the percent removals
for wastes having initial orthophosphate concentrations above 100 mg/1
were reasonably consistent irrespective of initial orthophosphate con-
centration. Considerable removals, 60-80%, were obtained with pH control
in the range of 9.0-9.5.
A number of dimensionless ratios were investigated to determine those that
could be used to predict removal and chemical dosages based upon waste
characteristics. The most sensitive parameters for all wastewaters were:
a) the chemical dosage per remaining total or orthophosphate concentra-
tion versus percent total or orthophosphate removal and b) chemical
dosage per initial calcium or total hardness versus percent total or
orthophosphate removal. Both of these parameters appeared to have use
in the design and operation of possible phosphate removal systems. The
parameters were reasonably sensitive for data from a number of different
wastewaters, for different phosphate concentrations, and for the three
chemicals investigated.
58
-------
No sludge production relationships were obtained from these laboratory
experiments. The majority of solids increase ratios were in the range
of 0.6-1.0 for duck wastewaters using all three chemicals.
Decisions on the most appropriate chemical for animal wastewaters are
difficult. The chemical of choice will depend upon the required dosage
and chemical cost and the costs of ultimate solids disposal. Some cost
comparisons were made to combine chemical costs and chemical demand
requirements. The costs of ferric chloride exceeded those of the other
two chemicals.
Although alum dosage requirements were less than those of lime for many
poultry wastewaters, the greater cost of alum resulted in cases when
the alum costs were equal to or greater than those for lime. For duck
wastewater, lime was the least cost chemical at all percent removals
followed by alum and ferric chloride in that order.
The costs of chemical precipitation of phosphates from poultry and dairy
manure wastewater were estimated using "average" wastes and possible
dilutions. Estimates at greater than 90-95% removals could not be
obtained using the predictive relationships developed in this study.
In this estimation, required alum dosages were less than for that of
lime. However, lime was found to be the least expensive due to lower
chemical costs. Chemical costs per 1000 gallons at the lowest dilution
explored, 1% TS, were of an order of magnitude or more greater than
those quoted for phosphate removal from municipal wastewaters.
Wastewaters containing duck processing wastes exhibited considerably
different phosphate removal characteristics and patterns than did wastes
from duck farms which contained no processing wastewater. These dif-
ferences underscore the fact that wastewaters from different sources
can have different phosphate removal characteristics and chemical demands.
The fact that data from many runs on wastewaters from similar sources
produced consistent results indicates that wastewaters from a specific
agricultural or industrial operation may have common removal charac-
teristics and chemical demands. Nevertheless, the constituents and the
concentration of a given wastewater havea definite impact on the choice
and effectiveness of the chemicals.
Although this phase of the project was directed toward chemical means
of removing phosphates from animal wastewaters, it should not be inferred
that this is the most effective method of phosphate control from these
wastewaters. The results of this project indicate that required chemical
concentrations are in proportion to the characteristics of the waste-
water, i.e., alkalinity, hardness, or phosphate. Ratios of chemical
dosages per initial orthophosphate concentration ranged up to 8-10 for
alum and lime at low, residual orthophosphate concentrations (less than
5-10 mg/1), and high orthophosphate removals (greater than 90%). Sludge
production may range between 0.5-1.0 mg/1 suspended solids increase per
mg/1 chemical used.
59
-------
To achieve low residual phosphate concentrations, a wastewater containing
100 mg/1 of orthophosphate may require about 800-1000 mg/1 of chemicals
which may produce an additional 400-1000 mg/1 of suspended solids for
ultimate disposal. More precise estimates of the chemical dosages
required for specific wastewaters can be obtained from Figures 12 and
13. Estimates of chemical demand and sludge production for wastewaters
of other characteristics can be obtained in a similar manner. The
large chemical demand and sludge production are decided disadvantages
to this method of phosphate control for concentrated animal wastewaters.
The general characteristics of animal wastes and wastewaters are such
that a high degree of treatment also will be necessary to remove BOD,
suspended solids, and other constituents if discharge to surface waters
is contemplated. Chemical precipitation of phosphates will add to
costs and operational problems. Approaches other than conventional
liquid waste treatment methods are needed for animal wastes.
•
Except for specific animal wastes or unique locations, animal production
facilities are located in areas where grass, crop, and brush land are
available. Land disposal of animal wastes and wastewaters offers a
reasonable alternative to the treatment of the wastes and discharges to
surface waters. With proper land and crop management, most of the phos-
phorus in applied wastewaters will be retained in the top few feet of
soil. Nitrogen in these wastes may be a problem and approaches to
nitrogen control will be discussed in other phases of this report.
Controlled land disposal should be considered as a high priority method
for phosphorus control from agricultural wastewaters because it is more
amenable to normal agricultural production operations, avoids the need
for chemical control and treatment plant operation, and eliminates
additional problems of chemical costs and sludge production. This
report indicates the types and magnitude of chemicals that may be
necessary and the magnitude of the solids disposal problem that may
result for phosphorus removal from animal wastewaters where discharge
to surface waters is practiced. The predictive relationships in the
report provide a method of meeting the effluent standards that might
be incorporated into discharge requirements for animal wastes to surface
waters.
60
-------
NITROGEN REMOVAL BY AMMONIA DESORPTION
INTRODUCTION
Ammonia derives its name from the tribe of people known as ammonians,
who lived in Libya, North Africa, about 2500 years ago. The compound
traditionally was prepared by using animal excreta (16). Systematic
studies since the eighteenth century have indicated the chemical compo-
sition of ammonia and the beneficial effects of ammonia and other nitro-
genous compounds on plant growth (17-19). A number of methods have
been utilized to meet the demand for nitrogenous fertilizers. Several
patents have been taken out for processes to recover ammonia from sewage
sludges (20). However, since the development of the Haber process for
synthesis of ammonia, the recovery of ammonia from wastes is no longer
commonly practiced. In recent years, the use of inorganic fertilizers
has steadily replaced the age-old practice of fertilizing soils with
animal wastes.
The nitrogen in the large quantities of animal wastes requiring dis-
posal has caused concern on its effect in the environment. Several
methods are available for reducing the nitrogen content of wastes such
as ammonia stripping, nitrification followed by denitrification, and
waste modification. Land disposal and crop management appear to be the
only methods which have been used to control the excess of nitrogen in
animal wastes. Little information exists on the other methods for
controlling discharge of nitrogenous compounds in animal wastes to the
environment. Studies with domestic sewage (21-26) indicated that nitro-
gen removal by ammonia desorption was feasible and that the principles
involved in the process may be applied to animal wastes.
Nearly half the quantity of nitrogen in the animal wastes may be in the
form of ammonia. Removal of ammonia by desorption may be a separate
unit process or combined with aerobic treatment methods. The feasibility
of removing ammonia by air-stripping is dependent on several factors
such as the concentration of ammonia in the waste, physical properties
of the waste, and the quantity of air.
THEORETICAL CONSIDERATIONS
This section outlines the equations and basic assumptions made in
developing the mathematical models used in this phase of the project.
Effect of Dissociation - The solubility of ammonia is very high. One
volume of water can dissolve as much as 670 volumes of the gas (16).
Only undissociated ammonia is available for desorption and in water it
exists in equilibrium with the ammonium ion:
NFL + H~0+ + NH/ + O (6)
O O "*~ H £
61
-------
This equation indicates that the amount of undissociated ammonia
in the system depends upon the hydrogen ion concentration. The equilib-
rium constant, K , can be obtained by applying the law of mass action
to this equilibrium.
[NH3] [H30+]
K (ammonia) = T (7)
eq [NH4+] [H20]
The ionization of water can be represented by:
H20 + H20 j H30+ + OH" (8)
Again, applying the law of mass action
[H30+] [OH']
K (water) = -^ «
eq [H20r
But since for practical purposes the value of [H20] is very nearly the
same, either in pure water or in dilute solutions (i.e., 55 moles HpO
per liter), a simplified expression can result:
Keq(water) . [H20]2 = [H30+] [OH"] = kw (9)
The equilibrium equation for the dissociation of ammonia, Equation 7, can
be simplified to the following again assuming [H20] to be a constant
under practical conditions:
x r n CNH4+] [°H"]
K (ammonia) [H^O] = —TMITT = k, (10)
The ratio of kw/kb can be expressed as:
[NHj] kb
&2
-------
] k
I = b . 10-PH where pH = -login[H..O+] (12)
5 w
Adding one (1) to both sides and rearranging the terms yields:
[NH3] 1 [NH3]
ammonia
-pH concentration
F" ' 10
w
F is the ratio of the undissociated ammonia divided by the total ammonia
concentration and is the fraction of ammonia nitrogen in the undisso-
ciated form. Rearranging Equation 13 yields:
= F (14)
10
pH
The undissociated (free) ammonia concentration can be obtained
by multiplying the total ammonia concentration (NHt as N) by 17F/14.
Equation 14 emphasizes that F is dependent upon pH and the iom'zation
constants of aqueous ammonia, k, , and water, k .
The values of k. and k vary with temperature (27). The data available
on k. and k at different temperatures indicate that these values increase
with increase in temperature (Figure 18). Mathematically the relation
between the value of k./k and temperature was determined to be:
kb/kw = [-3-39753 loge (0-024096)] x 109 (15)
where e is temperature in the Celsius scale.
The effect of temperature on the fraction of undissociated ammonia in
a water system is presented in Figure 19. Values of F for specific
temperatures can be found in the Appendix Table II.
63
-------
2.1
2.0
2.0i
Q
X
9
10
Q
X
.0
0)
'O
' 1.5-1
I.3J .5.
Kw
10
\
\
15 20 25
TEMPERATURE - °C
FIGURE 18
EFFECT OF TEMPERATURE ON THE
IONIZATION CONSTANTS FOR WATER AND AMMONIA
64
-------
From the Equation 14 or Figure 19, the following points can be made:
(a) When the pH is constant, the concentration of undissociated ammonia
increases with an increase in temperature. For example, at a pH value
of 9.6, about 45, 60, 68, 76 and 85 percent of ammonia is in the undis-
sociated form at 10, 20, 25, 30 and 35°C, respectively.
(b) When temperature is kept constant, an increase in pH results in an
increase in the concentration of undissociated ammonia. For example,
at 20°C about 60, 94, and 99 percent of ammonia is in the undissociated
form, at pH values of 9.6, 10.6 and 11.6 respectively.
(c) There is little change in the value of F at pH values either below
7.2 or above 10.6.
65
-------
(d) Over 94 percent of ammonia will be in the undissociated form at pH
values of 11, 10.6, 10.3 and 10.1 at 10, 25, 30 and 35°C respectively.
A knowledge of the quantities of alkali needed to keep a required level
of undissociated ammonia in a system is very useful. To change the pH
of a solution from any given point on the pH scale by two units, it will
require ten times more alkali to accomplish the second unitary rise.
For example, from (b) above it can be seen that at 20°C, the quantity
of alkali required for increasing free ammonia content of a solution
of an ammonium salt, in pure water, from 60 to 94 percent is only one
tenth of that needed to increase it from 94 to 99 percent. Jhe quan-
tity of alkali needed to obtain a known level of free ammonia will
decrease with increase in temperature.
The shape of the pH-F curve does not allow for easy reading of the F
value at all pH units. A linear transformation of the equation expressed
in Figure 19 would make its use more practical. Such a transformation
can be made by rewriting Equation 13 as follows:
(k /k )
(1-F) dissociated ammonia nitrogen _ v b' wy /16^
F undissociated ammonia lnpH
= 1°9lO(kb/kw) - PH (17)
Because k,/k is a constant at a given temperature, the relationship
(1-F)
between 1og,QP p ; ] and pH is linear. A semi-logarithmic plot of pH
versus (1-F)/F at different temperature would be a set of parallel
straight lines. This approach offers a graphical method for finding
the F values at different pH and temperatures (Figure 20) which avoids
errors inherent in drawing and interpreting sigmoid shaped curves. The
Appendix, Table III provides the relationship between F and (1-F)/F.
Transfer of Ammonia During Desorption - The desorption of ammonia from
wastewaters involves the contacting of a liquid and a gas phase in
various units such as spray towers, packed columns, aeration towers, or
diffused air systems. Whatever the mode of contacting, the gas and
liquid phases are brought together to transfer the ammonia. One phase
usually flows countercurrent to the other in a manner such that both
the phases are in contact. Ammonia being transferred from the liquid
to the gas phase must pass through the interface. It is difficult to
measure accurately either the length of the transfer path or the time
of contact. Even though an interface exists, its geometry is not well
defined. It can be assumed, as in heat transfer, that both phases are
separated and that transfer resistance layers are formed on either side
66
-------
10
50
100
( I-
FIGURE 20
GRAPHICAL PROCEDURE TO DETERMINE THE
FRACTION OF UNDISSOCIATED AMMONIA
-------
of the boundary. It is in these layers or films that the graetest
amount of resistance for mass transfer is encountered.
During the removal of ammonia from an aqueous solution by air stripping,
the greatest resistance to mass transfer occurs in the transfer from
the liquid to the gas phase. This resistance is due to the high solu-
bility and ionization of ammonia.
If we assume that under some conditions a steady transfer of ammonia to
air from the liquid takes place, then the rate of transfer is dependent
upon the concentration profiles of ammonia in both transfer resistance
layers, provided the following conditions are satisfied:
(a) laminar flow of fluids exists along the gas-liquid interface
(b) equilibrium at the gas-liquid interface is brought about instan-
taneously between the concentration of ammonia and the partial pressure
of the gas component.
The concentration profiles existing in the transfer of ammonia from the
liquid to gas phase are shown schematically in Figure 21.
If dN moles of ammonia are transferred across a surface area of A in
time dt, then according to Pick's law of diffusion,
dN = "D ' A ' Hx ' dt (18)
The resistance to diffusion can be assumed to reside in the two zones
(or films) MI and IN on each side of the interface I (Figure 21). The
liquid film resistance results in a concentration gradient from C (con-
centration of ammonia in bulk phase) in the liquid to C. at the inter-
face. Similarly, the gas film resistance results in a partial pressure
gradient from pi at the interface to p, the partial pressure in the bulk
gas phase. Due to the concentration and partial pressure gradients,
ammonia diffuses into the gas phase from the liquid phase.
The mass transfer coefficient for the gas film, kr, is defined by the
following equation: b
k - dN/dt MOI
K6 r (19)
Pi - P
If dN/dt is the rate of mass transfer in gm-mol/hr/cm2, and p. and p are
the above partial pressures expressed in atmospheres, the units of k
are gm.mol/hr - cm -atmosphere.
68
-------
INTERPHASE
ui
to
<£
I
CL
CO
<
CD
UJ
X
a
u*
o
UJ
o:
^
to
CO
UJ
K
Q.
<
i-
cc
GAS
FILM
DIRECTION OF MASS TRAMS
LIQUID
FILM
J
1
AC
± C,
JER FOR NH3 FROM LIQUID TO GAS PHASE
UJ
CO
o.
9
n
C)
UJ
x
t-
o
s
s
o:
2
U
O
^
O
o
M
FIGURE 21
SCHEMATIC OF THE TRANSFER OF
AMMONIA FROM A LIQUID TO A GAS PHASE
69
-------
The mass transfer coefficient for the liquid film, kL, is defined by the
following equation:
k =
(20)
The concentrations C. and C are usually expressed in gm.mol/liter, and
the units of kL are gm. mol/hr-cm2-gm. mol/liter, or cm/hr.
It is not always possible to measure the partial pressure and the con-
centration at the interface and it is necessary to employ transfer
coefficients based on the overall driving forces in the two films.
These coefficients may be defined by
P - P
(21)
=dN/dt_ (22)
L C - C
where KG and K, are the overall mass transfer coefficients in the gas
arid liquid phases, respectively; and P and C are the hypothetical
values at the interphase of the partial pressure of ammonia and the
concentration of ammonia which are in equilibrium with p and C,
respectively.
The magnitude of the resistance offered by each of the films is depen-
dent upon the solubilities of the substance. In the case of ammonia,
in view of its high solubility in water, the liquid film resistance is
negligible in comparison to the gas film resistance. Therefore, the
gas film resistance can be considered as the overall resistance for
mass transfer of ammonia from water to air (29) and KG equals kf.
It can be shown that HKL equals Kg where H is the Henry's constant.
Though either of these coefficients can be employed in studying mass
transfer operations, it is customary to use KG when the major resistance
to transfer is found in the gas film.
To design treatment facilities for desorbing ammonia from wastewaters,
it is necessary to know the mass transfer coefficients for ammonia
removal from the different types of wastewaters. Equations 21 and 22
describe the relationships affecting the mass transfer of ammonia
across an unit area of interface. We can presume that in an agitated
system, the surface is continuously renewed. Because it is very diffi-
cult to evaluate the interfacial area per unit volume, a, K. or 1C can not
L b
70
-------
be determined in practice. A combined desorption coefficient can be
determined and practically utilized. The combined desorption coeffi-
cients found in this study are denoted by 1C. The dimensions of Kn
are those of reciprocal time. u
Values of 1C can be determined in laboratory as well as full scale
systems. Trie numerical values of KQ found under these conditions are
system specific and may not be directly applicable to other conditions.
The factors that can influence the actual values of 1C include type,
size, and geometry of the aeration unit, type of diffuser or mechanical
aerator, characteristics of the waste, air flow rates, and temperature.
AMMONIA DESORPTION FROM ANIMAL WASTES
Individuals entering confined hog and poultry operations rapidly recog-
nize the strong smell of ammonia. The loss of ammonia from animal
wastes in these environments suggests that it may be possible to control
the removal of ammonia from these wastes.
From previous theoretical considerations (Equation 22) it can be seen
that with a specific desorption coefficient, the rate of removal, dN/dt,
will be increased if the concentration of ammonia in the waste, C, is
higher. Because animal wastewaters can contain high concentrations of
ammonia, ammonia desorption may be feasible with these wastewaters.
Animal wastes are defecated as a semi-solid material with a moisture
content of from 75 to 85 percent depending upon the species and the
nature of the diet of the animal. In many livestock operations, wastes
are added to a liquid system beneath the floor of the confinement building
In others, water is added to the wastes to facilitate hydraulic handling.
In still other operations, runoff water flowing over exposed wastes
creates a water borne waste.
The nitrogen in animal wastewaters will be in the form of organic and
ammonia nitrogen. In fresh wastes organic nitrogen is the predominant
form. In non-aerated holding tanks or pits, which are commonly used
prior to any treatment or disposal, microbial action will increase the
ammonia concentration in the wastes. In aerated holding units, such as
oxidation ditches, the degradation of organic nitrogen will occur with
nitrates and nitrites as the nitrogen end product if adequate residual
dissolved oxygen concentrations exist. Ammonia will tend to accumulate
if minimal aeration is practiced.
The nitrogen content, as N, of animal manure slurries has been indicated
to range from 0.3 to 1.3 percent for cattle; 0.2 to 0.9 percent for hogs;
and 1.8 to 5.9 percent for poultry (30). The total nitrogen content for
farm wastewaters has been shown to contain from 100 to 1812 mg per liter
for cowsheds and milking parlors (31).
71
-------
Runoff from a beef cattle feedlot can contain from 200 to 600 mg of
organic nitrogen per liter, and from 75 to 300 mg ammonia nitrogen per
liter (32). Another study indicated from 1 to 139 mg of ammonia nitrogen
per liter in feedlot runoff with an ammonia nitrogen to Kjeldahl nitrogen
ratio of 0.01 to 0.04 (33). The higher ammonia concentrations occurred
during the summer and fall.
Aeration towers are used to degasify ground waters and to remove ammonia
from wastewaters at municipal waste treatment facilities. Towers could
be used with animal wastewaters. The installation of the towers and
pumps, the inherent scaling and clogging that might occur with high
strength organic wastes would add to the costs and operational problems
at an animal production facility. Aeration towers do not appear to be
the most feasible approach with concentrated animal wastes.
Since holding tanks are in common use at animal production facilities,
it appears logical to consider diffused aeration systems for ammonia
stripping. In addition to ammonia removal, these systems would reduce
odors and accomplish some degree of aerobic biological treatment. Unlike
aeration towers, additional pumping equipment and towers would not be
needed and scaling and clogging should not be a problem. Both aeration
tower and diffused aeration systems would require an air supply and dis-
tribution system which do not form part of the existing animal pro-
duction facilities.
The important factors governing the release of ammonia during aeration
of animal wastewaters are: (a) the quantity of ammonia available in a
form suitable for molecular exchange (24-26); (b) diffusivity of ammonia
across the gas-liquid interface (22) and (c) rate of aeration (23, 24).
The influence of these and other factors on the desorption of ammonia
from animal wastes is discussed in this report.
OBJECTIVES AND METHODS
The objectives of this study were to: a) determine the desorption coef-
ficient, KD, for poultry and dairy manure wastewaters, b) develop pre-
dictive relationships involving the factors governing the desorption of
ammonia and their applicability to actual systems, and c) verify the
relationships in pilot plant studies.
Mathematical Approach - The amount of ammonia desorbed from its solution
into air is directly proportional to the concentration of ammonia in the
liquid, interfacial area of exposure, time of desorption, temperature,
and atmospheric pressure. The following relationship exists between
the quantity of ammonia lost per unit area of interface (AC1) and a
mass transfer coefficient.
AC1 = k'CAt (23)
where k1 is a constant at a given temperature and pressure.
72
-------
If the total area of interfacial surface of a volume of liquid is A.,
then the change in the ammonia concentration in the liquid due to desorp-
tion (AC) from the entire surface in the duration At is:
AC = k1 • A1 . C - At (24)
This equation indicates that greater quantities of ammonia can be desorbed
by increasing the time of exposure, area of exposure, and the concentra-
tion of ammonia in the liquid.
When air is bubbled through the liquid at a fixed rate, the total inter-
facial area is the sum of all surface areas of the air bubbles. Increas-
ing the rate of air flow through a given volume of liquid increases the
number of air bubbles traveling through the liquid. If "n" air bubbles
of surface area A, are formed when an unit volume of air is bubbled
through an unit volume of liquid, then the total interfacial surface
area formed is equal to nAb- The quantity of ammonia desorbed from a
diffused air system is dependent upon the rate of air flow and the size
of air bubbles formed. With a given diffuser system the total inter-
facial area is directly related to the rate of aeration.
The amount of ammonia removed from a liquid when a continuous stream of
air bubbles passes through it depends upon the following:
a) concentration of free ammonia in the liquid which is a function of
pH and temperature
b) rate of air flow
c) volume of liquid
d) mass transfer coefficient
e) duration of desorption
f) bubble contact time
Batch Desorption Systems - Using previous relationships and the above
factors, an equation can be developed for batch desorption systems:
•= -k1 • • C • F (25)
where k1 is the mass transfer coefficient; F is the proportion of free
ammonia; C is the concentration of total ammonia in the liquid; S is the
rate of air flow, and V is the volume of liquid. As noted earlier, the
73
-------
mass transfer coefficients in actual systems will be dependent upon the
desorption systems that are used.
The product of k1 and S/V is the desorption coefficient, KQ for the
system. Therefore, Equation 25 can be written as:
{= -KU •
dt (26)
or
(27)
where C, and C2 are the concentrations of total ammonia at times t-j and
tp, respectively.
Continuous Desorption Systems - If V is the volume of the liquid in the
unit in which the desorption of ammonia is carried out, and Q is the
rate of flow of the liquid into the unit, Q is the rate of outflow,
assuming no losses due to evaporation. If C, and C2 are the concentra-
tions of ammonia in the influent and effluent, respectively, then a mass
balance on the unit at equilibrium will show:
Q • ( - C) = K • F • C • V (28)
2 ,j • • 2
or
- C)/C = K • F • (V/Q) = K • F . t (29)
22 D p • . H
where t,, is the average liquid retention time in the unit and the average
time of desorption in a continuous flow system. Thus, for whatever
decrease in ammonia that may be required, the liquid detention time for
which a system should be designed may be determined if 1C and F are known.
F is dependent upon pH and temperature and can be determined from Figure
20. KD is dependent, among other things, upon the nature of the liquid
and the rate of aeration, and again is sytem dependent.
Variable pH - Equations 27 and 29 are valid for systems where F is constant.
To examine the systems in which F is not constant due to pH changes, modi-
fications of the equations are necessary. When a solution of an ammonium
salt is made alkaline, ammonia is converted into the undissociated form
74
-------
+ NaOH
.OH + NaCl
(30)
NH4OH
(31)
These are first order reactions and the rate of decrease in the concen-
tration of ammonium ion is logarithmic with time. If the pH is not
maintained constant, there will be a proportionate reduction in the pH
level of the system. In these systems, the rate of decrease in pH will
be linear.
PHt = PHo - z • t
(32)
where pH. and pH are the values of pH at times t and o, respectively,
\f \J
and z is a constant and the slope of the pH-time curve. An example is
shown in Figure 22.
SYMBOL
X
Q.
30 60 90 120
PERIOD OF AERATION - mins.
FIGURE 22
REDUCTION OF pH DURING AMMONIA DESORPTION
It would be useful to develop relationships between pH, time, concen-
tration of total ammonia in the liquid, and KQ for predictive and design
purposes. Since F is a function of pH, Equation 26 can be rewritten as
follows using Equation 14:
75
-------
F • dt = -Kr
10
pH
10h
dt
(33)
When the rate of aeration and temperature are constant, the values of
1C. and k./k do not vary. Under these conditions when pH is allowed
to fall as ammonia is desorbed, then C, pH and t are the only variables.
Substituting for pH in Equation 33 with Equation 32, it follows that:
dC
C
- k'
~ "KD
,QpH - zt
inP^n" Z^ 4. (If /!/ \
JO o + (kb/kw) _
dt
(34)
If pH-, and C, are the pH and concentration of total ammonia at time t,
and pHL and C2 are the pH and-concentration of total ammonia at time t?,
upon integration it follows that:
pH,
(35)
Designating loge(10pH+ Kb/Kw) as L, and using appropriate subscripts for
L to denote the values at different pH, we can condense this equation to:
log.
z • Ioge10
(36)
Equation 32 can be written as:
z =
(t2-t^T
(37)
Substituting this for z in Equation 36 and rearranging the terms the
following equation is obtained:
76
-------
[pH2-pHj [log (C2/C,)] [loglO]
I/ _ C. c C. \ 6 I in\
KD (t2 - V (L^) (38)
A tabulation of the values of L at different temperatures and different
pH values are presented in the Appendix, Table IV.
Equation 38 was used in all diffused aeration experiments where-pH was
not controlled. Equation 27 was used in all diffused aeration experiments
where the pH was controlled to specific levels. An example of how the
desorption coefficient, 1C, was determined in experiments where the pH
and temperature both varied is presented in the Appendix, Table V.
Aeration Towers - Previous equations in this Section can be used directly
with diffused aeration systems. Additional mathematical relationships
are required for aeration towers.
The conventional aeration tower consists of a cylindrical or rectangular
vessel with a device to spray liquid from the top, an air injection
system or natural convection to have air enter the bottom and two openings,
one at the top to allow exhaust gases to escape and one at the bottom to
withdraw the liquid from the tower. The tower is packed with different
types of materials to obtain a large increase in the surface area of
contact.
In these towers, the amount of ammonia desorbed from the liquid depends
upon the following: a) rate of downward flow of water (f. -gm/sq cm/hr);
b) rate of upward flow of air (fp-gm/sq cm/hr); c) height of the tower
(x cms); d) initial concentration of ammonia (gm/liter); and e) the
mass transfer coefficient, K. (cm/hr).
A cross sectional area of 1 sq cm with a flow of liquid (f, ) counter
current to an air flow (f-) can be used for descriptive purposes (Figure
23). A concentration gradient will be established in the tower whereby
ammonia is removed from the liquid and transported out of the system by
the air. If the concentration of the free ammonia in the influent and
the effluent liquid is C, and C? respectively and the concentration of
free ammonia in the incoming and the outgoing air is p-, and p2, respec-
tively, a material balance for the system will result in:
fL(Cl - C2) = fG(p2 - P!> (39)
77
-------
AX
C+ AC
LIQUID f,
p + Ap
AIR f.
X = HEIGHT
FIGURE 23
MASS TRANSFER DURING COUNTERCURRENT
AIR AND LIQUID FLOW
78
-------
The rate of transfer of ammonia from the liquid to the gas phase across
the interfacial boundary between the liquid and gas phase in an infini-
tesimal height, dx, of the column is:
fL • dC = fe • dp = K,_a • p(C. - C) dx (40)
where p is the density of the liquid; a is the interfacial surface area
in unit volume of the column, and C. is the concentration of ammonia in
equilibrium with the gas phase at the interface. Equation 40 can be
rewritten as:
c4r w
' p -i
Since the ratio in brackets is constant for a given set of conditions,
the total change in the concentration during flow through the column
can be obtained by integration. The term in the brackets in Equation
41 has the dimension of length. This term is sometimes referred to as
the height of a transfer unit (HTU). In practical use of the equation,
the term K.a is synonymous with the desorption coefficient, 1C. It must
be noted that this equation is valid if there is no packing in the column.
If the column is packed then it becomes necessary to apply correction
factors which depend upon the properties of the packings to the desorp-
tion coefficient.
To observe the relationship between the height of the column, x, and the
change in the ammonia concentration, Equation 41 can be rewritten as:
AX = -HTU • |£- (42)
AX
Cl/C2 - eHTU (43)
An increase in the height of the column will exponentially decrease the
concentration of ammonia in the liquid.
MATERIALS AND METHODS USED IN THE STUDY
Materials - Experiments on desorption of ammonia were conducted using the
following liquids:
a) solutions of different concentrations of ammonium chloride in water
b) suspensions of poultry manure in tap water
79
-------
c) suspensions of dairy manure waste in tap water
d) mixed liquor from an oxidation ditch treating poultry wastes
The samples of poultry and dairy manure used in these experiments were
collected from facilities at Cornell University. The oxidation ditch
was located in the Agricultural Waste Management Laboratory at Cornell.
Methods - To correlate the different factors that may influence desorption
of ammonia from wastewaters, and characterize the wastewaters, the fol-
lowing analyses were made: (a) all forms of nitrogen; (b) pH value;
(c) total solids; (d) chemical oxygen demand (COD); (e) surface tension;
and (f) viscosity.
Total solids, total Kjeldahl nitrogen, ammonia, nitrite and nitrate
nitrogen were determined by the methods described in the Standard Methods
(12). Chemical oxygen demand (COD) of the samples was determined by the
rapid method (35). pH value of the samples was measured using a pH meter.
In some experiments, biochemical oxygen demand (BOD) and phosphorus con-
tents of the samples were determined by the methods described in Standard
Methods.
Surface tension at the air-liquid interface was measured by using the
method of duNouy, modified by Harkins and Jordan (36, 37). To measure
the viscosity of the liquids, an Ostwald constant volume flow capillary
viscometer (38) was used. To separate the particulate matter in the
waste suspensions that may have clogged the capillary, these suspensions
were filtered through tissue paper before the analysis. Viscosity of
some of the samples also was measured by a Brookfield rotational vis-
cometer (39). For measurements with this viscometer, separation of the
particulate matter was not necessary.
Experimental Setup - Experiments were conducted with an aeration tower
apparatus as well as with diffused air systems.
a) Aeration Tower - The experiments were conducted in a plastic tower
approximately 6 inches inside diameter filled with plastic Raschig rings
up to a height of four feet. The tower, auxilliary equipment, and the
plastic media are shown in Figure 24. The 1/2" long Raschig rings were
cut from hard plastic tubing having an outer diameter of 1/2" and inner
diameter of 3/8". These were made available to the project by the
Chemical Engineering Department at Cornell. Because of the small size
of the tower, small media were necessary to minimize wall effects The
media size to tower diameter was 1:11.5, above the value of 1-8 generally
used as a guide to minimize wall effects. The media were dumped into
the tower, and the tower shaken a few times to achieve random placement
of the media.
The diffuser and the media distributed the liquid throughout the tower
A visual test using methyl orange was used to check the distribution. '
80
-------
EXHAUST
DISTRIBUTOR
ROTAMETER
COLLECTION
BOTTLE
AIR
COMPRESSOR
PLASTIC MEDIA
USED IN AERATION
TOWER
FIGURE 24
EQUIPMENT USED IN AERATION
TOWER EXPERIMENTS FOR
AMMONIA STRIPPING
81
-------
Whenever a different liquid was used, the tower was first flushed with
the new liquid before the experiment was begun.
b) Diffused Aeration - Small and large scale experiments were made on
batch and continuous flow units with diffused aeration systems.
i) Bench Scale Setup - Batch Units - Studies were conducted in the
laboratory in both a 2 inch inner diameter plastic column with air
entering through a diffuser at the bottom and in beakers or tall glass
cylinders. The equipment is shown in Figure 25 for experiments using
the beakers. The air was supplied through a rotameter, pressure relief
TYGON
TUBING
AIR
COMPRESSOR
VESSEL CONTAINING
LIQUID
DIFFUSER
FIGURE 25
SCHEMATIC OF BENCH SCALE - BATCH STUDY
EQUIPMENT FOR AMMONIA DESORPTION
valve, and water saturation flask. The air was saturated to avoid exces-
sive loss of moisture in the experiments. Most of the runs were made at
room temperature, 20-23°C. In some runs, temperature of the system was
kept constant at specific levels by immersion of the units in a constant
temperature water bath.
82
-------
ii) Bench Scale Setup - Continuous Flow Units - A constant flow into
the desorption unit was maintained with the aid of an "electrolysis
pump" (40) to obtain a constant detention time in each experimental
run. As in the case of batch units, air was monitored through a rota-
meter, pressure relief valve, and water saturation flask.
A battery of continuous flow units was utilized by connecting a number
of cylindrical vessels of different dimensions in a series (Figure 26).
AIR FLOW METERS
OUTFLOW
LIQUID LEVEL
INFLOW
TYGON TUBING
FIGURE 26
SCHEMATIC DIAGRAM OF THE CONTINUOUS
FLOW LABORATORY EXPERIMENTS FOR
AMMONIA DESORPTION
By adjusting the height of the outlet on the last vessel, different
detention volumes could be obtained. By this setup it was possible to
examine, in the same run, the effect of the following on the desorption
of ammonia from the liquid: detention time, rate of aeration, pH, and
concentration of ammonia.
iii) Pilot Plant Setup - Batch Units - This setup was similar to the
laboratory bench scale setup. The desorption tank was a 7' high
83
-------
cylindrical vessel of 3' diameter (Figure 27). The j]^ diffused
through a 3' long perforated pipe. The maximum air flow rate the
obtainable was 30 SCFM.
FIGURE 27
PILOT PLANT EQUIPMENT FOR BATCH AND
CONTINUOUS FLOW AMMONIA DESORPTION STUDIES
iv) Pilot Plant Setup - Continuous Flow Units - The desorption tank was
a high cylindrical vessel of 3' diameter and had outlets at different
heights. Different detention volumes could be obtained by proper choice
of effluent outlet height. Constant flows of the liquid and a solution
of sodium hydroxide were maintained with the aid of two peristaltic pumps.
The ammonia-rich manure suspension was stored in a 900 gallon tank and
was gently aerated to keep solids in suspension. The change in ammonia
concentration was negligible in this tank. The suspension was pumped at
different rates to verify the laboratory results and to extend the
application of previous data.
84
-------
RESULTS
General - The effect of the following factors on the removal of ammonia
from animal wastewaters was studied as appropriate in tower and diffused
aeration systems:
a) concentration of ammonia
b) PH
c) rate of air flow
d) rate of liquid flow
e) temperature
f) concentration of solids
g) viscosity of the liquid
h) surface tension of the liquid
The effect of ammonia desorption on the other parameters such as BOD,
COD, total Kjeldahl nitrogen, phosphates, and total solids was deter-
mined in some experiments. The number of experiments and the nature
of the study is summarized in Table 7. The desorption coefficients
obtained in this study should be recognized as system specific. Extrap-
olation of these values to other situations should be done with care.
Small Scale Studies
Aeration Tower - The liquids flowed through the column at controlled
rates and samples of influent and effluent were analyzed.
The parameters investigated included the following:
(a) rate of flow of liquid (0.5, 1.0, and 1.5 liters per minute)
(b) rate of aeration (1.85, 2.0, and 4.5 cfm.)
(c) concentration of total ammonia in the liquids (approximately 100,
500, and 1000 mg NH3~N per liter).
All the experiments were conducted at 20-23°C. To adjust the pH levels
of the samples, sodium hydroxide and calcium hydroxide were used with
solutions of ammonium salts and suspensions of poultry waste in tap
water, respectively. All of the experiments conducted in this subphase
were conducted at a constant pH.
The pH of the effluent was continually adjusted to the desired level
after each pass through the aeration tower and before it was again placed
85
-------
TABLE 7
SUMMARY OF AMMONIA DESORPTION EXPERIMENTS
Experiment
Number of
Runs
Materials Used
00
CTi
LABORATORY SCALE
Aeration towers
Diffused aeration
constant pH
pH not controlled
Continuous flow,
diffused aeration
25
18
25
51
17
27
12
•ammonium chloride in tap water
-poultry waste
-ammonium chloride in tap water
-poultry waste
-ammonium chloride in tap water
-poultry waste
-poultry waste treated by oxidation ditch
-dairy waste
-ammonium chloride in tap water
-poultry waste
PILOT PLANT FACILITY
Diffused aeration,
batch process
Diffused aeration,
continuous flow process
20
18
-ammonium chloride in tap water
-poultry waste
-poultry waste
-------
in the feed bottle for another pass. Eight to ten passes were made
with each sample. The ammonia in the influent and the effluent of the
liquid used in each pass was determined.
The objective of this study was to find whether the desorption of
ammonia at constant pH values followed a first order reaction and
whether pH and flow rate had any effect on the rate of desorption.
Typical results obtained with suspensions of poultry wastes and solu-
tions of ammonium chloride are given in Figures 28 and 29. The results
suggest that the rate of ammonia removal followed the typical first order
reaction indicated by Equation 27. Equation 27 and Figure 28 were used
to determine the coefficient of desorption for the different experimental
runs.
The effect of initial ammonia concentration of the liquids, rate of
liquid flow, and rate of aeration is noted in Figures 30 and 31. In
these runs the air flow rate was constant at 4.5 ft.3/minute and the
gas to liquid molar ratios were 0.06 at the liquid flow rate of 0.5 1pm,
0.12 at 1.0 1pm and 0.18 at 1.5 1pm. The initial total ammonia content
of the liquids did not appear to affect the rate of desorption (Figure
30). The rate constants for the poultry waste suspensions were in the
same range as that of the solution of ammonium chloride in tap water.
The results also indicated that Equation 27 was valid for high concen-
trations of ammonia in animal wastewaters.
The rate of liquid flow was found to affect the rate of removal of
ammonia (Figure 30). By increasing the rates of liquid flow, when the
aeration rate was the same, it was observed that the rate of removal
of ammonia, i.e., 1C, decreased. Higher rates of liquid flow permitted
less time of contact between gas and the liquid phase. The changes in
the 1C value did not appear to be linearly related to the rate of liquid
flow. As indicated earlier, the rate of ammonia desorption depends upon
the rate of air flow through the liquid.
Since both the rate of liquid flow and rate of aeration were found to
affect the value of 1C, the interrelationship between these two factors
was examined. By increasing the rate of aeration, keeping a constant
rate of liauid flow, an increase in the rate of desorption of ammonia
was noted (Figure 31). In these runs, the liquid flow rate varied from
0.5 to 1.5 liters per minute. The gas to liquid molar ratio was 0.027
for one ft.3of air per one liter of liquid thus permitting the gas to
liquid ratio at other air:liquid flow ratios noted in Figure 31 to be
obtained directly. The results of these experiments indicated that the
value of 1C was unaffected by an air to liquid flow ratio less than 4.
Above this value an increase in the value of 1C was noted.
Diffused Aeration Studies
a) Constant pH - One liter samples of tap water containing varying
amounts of either ammonium chloride or poultry wastes were used in
87
-------
60O,
o»
I
UJ
o
o
o
O 100-1
<
50
AIR FLOW
4.5 cfm
WATER FLOW
0.5 Ipm
2468
PASSES THROUGH TOWER
10
FIGURE 28
AERATION TOWER EXPERIMENTS
POULTRY WASTEWATER
DIFFERENT pH LEVELS
88
-------
20CH
(A
O
100
1
o
K
tr
h-
g 50-
0
z
o
0
z
2
20-
C
>v ° ^""^o
^0 ^^x ^^o^.
\0 OX^PH 10 ^^^0--^
°v
°>DH II
o
AIR FLOW
4.5 cfm WATER FLOW
0.5 Ipm
) 2 4 6 8 10
PASSES THROUGH TOWER
FIGURE 29
AERATION TOWER EXPERIMENTS
TAP WATER PLUS
C1
DIFFERENT pH LEVELS
89
-------
.25,
.20-
i
/ J
X0
0.5 Ipm — /-v ,_,,,_
o LIQUID
FLOW
RATE
1.0 Ipm /
•0
o/
200 400
1.5 Ipm
600 800 \OCX
INITIAL AMMONIA CONCENTRATION-mg/l
FIGURE 30
AERATION TOWER EXPERIMENTS
VARIATION OF KQ WITH INITIAL AMMONIA
CONCENTRATION AND LIQUID FLOW RATE
90
-------
«t
V)
o
Q.
.25,
.20
.15-
« .KM
.05
0
AIR
FLOW
23456
AIR FLOW RATE-cfm
2468
AIR:LIQUID FLOW RATIO - ft3/liter
8
10
FIGURE 31
AERATION TOWER EXPERIMENTS
VARIATION OF Kn WITH AIR FLOW
RATE AND AIR:U LIQUID RATIO
91
-------
these batch study experiments. The choice of this volume of the liquid
was made so that all aeration rates could be directly related to the
liquid volume. Foaming occurred when suspensions of poultry wastes were
aerated. Dow Corning si li cone antifoam was used to reduce foaming, in
separate experiments, Kp was found not to be affected by the silicone
antifoam at the concentrations used.
The effects of the following parameters were studied: (a) concentrations
of total ammonia from 50 to 1000 mg as N per liter; (b) values of pH
from 9 to 11; and (c) aeration rates from 0.5 to 2.5 cfm. These rates
reflect the uncorrected rates of air flow. Subsequent data used in the
report are air flows corrected to standard conditions. Because the
volume of the liquid was small, automatic pH control equipment available
in the laboratory could not be used. The pH value of the liquid was
readjusted periodically to the desired level throughout a run. This
occurred about every 10 to 15 minutes, especially for the runs in which
pH value of the liquids was high.
The results of these batch studies conducted at constant pH values indi-
cated that the decrease in ammonium nitrogen concentration with time
followed a first order reaction (Figures 32 and 33). The desorption
coefficient did not appear to vary with initial total ammonia nitrogen
concentrations in tap water (Figure 34) substantiating the data obtained
in the aeration tower experiments. However, the results of some of the
experiments with poultry waste (Figure 34) suggested a change in the
value of 1C with a change in the initial total ammonia concentration
of the liquid. This difference would appear to be due not to the initial
total ammonia content but to the other constituents present in poultry
wastewater.
b) Uncontrolled pH - No attempts were made to adjust the pH value in
these batch study experiments. Varying concentrations of ammonium
chloride in tap water, suspensions of poultry waste and dairy manure
waste, and mixed liquor from an oxidation ditch treating poultry wastes
were used.
The effects of the following parameters were examined: (a) rate of
aeration; (b) temperature; and (c) total solids content. In some of the
experiments, the liquids were analyzed for COD, viscosity, and surface
tension.
The experimental procedure in these runs consisted of analyzing samples
taken from the aeration units at specific intervals for the following:
(a) temperature; (b) total ammonia; (c) pH, and (d) total solids.
b-i) Effect of Temperature - During aeration the temperature of the
liquid decreased (Figure 35) and it was noted that the value of 1C also
decreased. Similar curves were obtained with poultry manure wastewater,
dairy manure wastewater, and oxidation ditch mixed liquor suspensions.
92
-------
lOOOi
BATCH STUDIES
CONSTANT pH
50
20 40 60
AERATION TIME-minutes
FIGURE 32
DIFFUSED AERATION EXPERIMENTS
TAP WATER PLUS NH4C1
DIFFERENT pH LEVELS
93
-------
2001
(0
O
I1
i
Z
g
i-
UJ
O
Z
O
<
O
2
100
50-
BATCH STUDIES
CONSTANT pH
AIR FLOW-2.44 cfm
10
20 40 60 80
AERATION TIME - minutes
100
FIGURE 33
DIFFUSED AERATION EXPERIMENTS
POULTRY WASTE WATER
DIFFERENT pH LEVELS
94
-------
2.5-,
2.0-
| 1.5
k.
«• i.oj
Q
.5-
BATCH STUDIES
CONSTANT pH
POULTRY
WASTE o
AIR FLOW
2.4 cfm
TAP WATER + NH4CI
AIR FLOW-1.1 cfm
200 400 600 800
INITIAL AMMONIA CONCENTRATION-mg/l
FIGURE 34
DIFFUSED AERATION EXPERIMENTS
VARIATION OF K WITH
INITIAL AMMONIA CONCENTRATION
95
-------
AMMONIUM CHLORIDE SOLUTIONS
UNCONTROLLED pH
6 SCFH/liter
PERIOD OF AERATION - hrs.
FIGURE 35
REDUCTION OF LIQUID TEMPERATURE DURING
AMMONIA DESORPTION
With higher rates of air flow, greater reductions in the liquid tempera-
tures occurred. To determine the effect of temperature on the rate of
desorption of ammonia by. aeration a specific series of experiments was
conducted. The liquids used in these experiments were the same as those
noted under b) above and were maintained at a constant temperature using
a water bath. The experiments were conducted at about 10, 15, 20, 25,
30, and 35°C and the values of Kn were determined at air flow rates of
6, 10, 12, and 20 SCFH per liter.
The results of these experiments (Figure 36) indicated that with an
increase in either temperature or rate of air flow through the system,
there was an increase in KD. The intercepts of the air flow-KD lines
have a positive intercept on the x axis indicating that ammonia desorp-
tion does occur even when there is no aeration. The value of this inter-
cept is an estimate of the rate of ammonia loss that could occur under
quiescent conditions in bodies or containers of water and wastewater.
96
-------
IT
UJ
t I 5-
:n
u.
o
co 10-
I
5-
.2
I \
,4 .6
KD PER HOUR
(a)
1,0
35-
30-
6 SCFH/L
°
DC
a:
UJ
Q. 10-
2
UJ
I-
20 SCFH/L
I
.2
i
.4
i
.6
KD PER HOUR
(b)
i
.8
FIGURE 36
EFFECT OF TEMPERATURE AND AIR FLOW
ON THE RATE OF DESORPTION OF AMMONIA FROM
ITS SOLUTION IN WATER
1.0
97
-------
In subsequent experiments to verify the 1C values under quiescent con-
ditions, ammonia solutions in different containers with varying surface
exposure areas were used. Slow desorption took place and the extent of
ammonia loss was directly proportional to the surface area exposed
(Figure 37).
The desorption coefficients obtained under quiescent conditions (Figure
37) and by the intercept of the air flow-IC lines (Figure 36) were approx-
imately three times the rate expected baseH on the diffusivity of ammonia
alone (2.6 x ID'4 sq ft per sec at 25°C). The coefficients obtained in
this study thus include some small amount of surface agitation of the
liquid due to normal air movement as well as simple diffusion.
The intercepts of the temperature-Kp lines (Figure 36) have a positive
intercept on the y axis indicating that the desorption of ammonia can
occur only at temperatures greater than 3-5°C. The slope of these lines
provides an estimate of the effect of temperature on the efficiency of
ammonia desorption and the ability to predict the effect of changes in
ambient conditions on the process.
b-ii) Effect of Surface Tension and Viscosity - It was observed that the
surface tension and viscosity of the liquids were lowered by raising the
temperature of the liquids. Typical data from experiments using poultry
manure wastewater are presented in Figure 38. There was no direct cor-
relation between the surface tension and 1C at different temperatures.
However, at a given rate of air flow, the ratio of 1C at two given temp-
eratures was almost equal to the square of the ratio of the differences
of surface tension between the sample and the surface tension of a
solution of ammonium chloride of equivalent total ammonia concentration
in water (Table 8).
A comparison of the viscosity at different temperatures and the corres-
ponding values of Kg at a given rate of aeration indicated a linear
relationship on a log-log plot (Figure 39). The data presented in
Figure 39 was obtained over a temperature range of from 10-30°C. With
an increase in the viscosity there was a proportionate decrease in 1C.
This relationship can be expressed as: "
KD = a(y)b (44)
where y is the viscosity of the liquid, b is the slope of the lines in
Figure 39, and a is the value of KQ when y is one centipoise. A com-
parison of the slopes, b, (Table 9) indicated that neither the type of
liquid nor the air flow rate within the range of 6-20 SCFH/liter of
liquid had a significant effect. A slight difference in this relation-
ship was noted at an air flow rate of 3 SCFH/liter
98
-------
0.6-
0.4-
cc
D
O
I
0.2-
NH4 Cl IN
TAP WATER
37°C
20 C
1 1 1 1
0 5 10 15 20
AREA OF EXPOSURE-squore centimeters
FIGURE 37
Kn VARIATIONS DUE TO TEMPERATURE
U UNDER QUIESCENT CONDITIONS
99
-------
60-,
Ul
UJ
UJ
0 50-1
tr
UJ
a.
to
UJ
a 4.0-
IO°C
SURFACE TENSION
T
T
20°C 30° C
TEMPERATURE
40°C
2.5n
2.0H
UJ
to
o
a.
i-
z
UJ
o
I.5H
1.0-
IO°C
VISCOSITY
20°C
30°C
TEMPERATURE
40°C
FIGURE 38
EFFECT OF TEMPERATURE ON
SURFACE TENSION AND VISCOSITY OF
SUSPENSIONS OF POULTRY WASTE IN TAP WATER
100
-------
Temperature °C
'1
10
10
20
20
20
26
26
'2
20
36
26
31
36
31
36
TABLE 8
CORRELATIONS* BETWEEN KQ AND SURFACE TENSION
1.54
2.72
1.11
1.61
2.17
1.36
1.67
Ratios of difference
in surface tensions
1.19
1.72
1.13
1.27
1.45
1.12
1.28
**
Square of the ratios
1.42
2.98
1.28
1.61
2.10
1.26
1.63
for suspensions of poultry manure and for solutions of ammonium chloride in tap water
**
At each temperature, the surface tension of the waste and an equivalent solution of ammonium chloride was
determined, i.e., a
.,,, -,
- a
, .
A ratio was then made of the differences at the two temperatures, i.e.,
-------
1.0'
.8-
.6-
NH4CI
POULTRY WASTE
DAIRY WASTE
3
O
.4-
,2
*- 3 SCFH/L
• - 6 SCFH/L
o-10 SCFH/L
• -12 SCFH/L
a -20 SCFH/L
OXIDATION
DITCH LIQUOR
POULTRY
DAIRY
2 4
VISCOSITY - centipoises
6
FIGURE 39
EFFECT OF VISCOSITY ON
AMMONIA DESORPTION RATE
102
-------
TABLE 9
THE VALUES OF a AND b FOR THE EQUATION KD = ayb
AIR FLOW RATE .
ID (CFH/L) a b r
Poultry manure 3 0.21 -2.7 0.96
wastewater 4 0.44 -2.9 0.94
6 0.85 -3.4 0.92
10 0.91 -3.7 0.88
12 1.06 -3.5 0.95
13.3 1.16 -3.9 0.87
16 1.37 -3.5 0.93
20 1.68 -3.5 0.95
Dairy manure 6 0.18 -3.3 0.89
wastewater 12 0.80 -3.6 0.88
20 1.98 -3.5 0.92
Water plus 6 0.98 -3.4 0.91
ammonium 10 1.28 -3.3 0.88
chloride 20 1.79 -3.2 0.82
*
r is the coefficient of correlation for the least squares fit for the above
equation.
NOTE: The above data were obtained from batch experiments in which no
attempt was made to control the pH. Initial NHj-N concentrations
varied from 100 to 2000 mg/1. The volume of the liquid that was
aerated varied from 250 ml to 1000 ml.
103
-------
The fact that Equation 44 was consistent over many air flows, with dif-
ferent wastes, and over a broad temperature range from 10°C to 35°C
suggests its usefulness for general design purposes with wastes of dif-
ferent characteristics.
b-iii) Effect of Solids Content on Viscosity - The viscosity and total
solids content of samples containing varying amounts of poultry waste in
tap water were determined. The results indicated that viscosity of the
suspensions increased with increase in the solids content and therefore
it was expected that with an increase in the solids content of a waste,
the value of KD would decrease (Figure 40). Thus with all other con-
ditions remaining constant, higher air flow rates are required to accom-
plish a given degree of ammonia removal with an increase in the solids
content.
The data presented in Figure 40 were obtained from batch experiments con-
ducted at 20°C. No attempt was made to control the pH. The straight
lines shown in Figure 40 represent the least square fit of the available
data. Due to the variability of the data at low air flow rates, little
inference should be made of the fact that there appears to be a cross-
over of the lines at those rates. The value of the Figure is to illus-
trate the difference in KQ as related to the total solids content of the
liquid.
Quality of Effluent from Ammonia Stripping Systems - The treatment of
wastes to desorb ammonia also alters the concentration of other waste
characteristics. The addition of a base to waste solutions can precipi-
tate colloidal material reducing a portion of the organic and inorganic
content. The range of removals of other constituents that occurred in
these systems is illustrated in Table 10. The data were obtained from
runs in which the pH was not controlled.
The data shown were obtained by comparing the analysis of the untreated
waste and the supernatant of the wastes after treatment. The decreases
in COD, total nitrogen, and orthophosphate are the result of the pre-
cipitation of the colloids and suspended matter that took place. In
some runs with sodium hydroxide, coagulation and sedimentation did not
occur and the high amounts of base needed resulted in an increase in
the total solids content.
This information indicates that removal of considerable amounts of other
contaminants will occur when ammonia is desorbed from waste solutions.
In general, when ammonia removals were high, the removals of the other
constituents also were high.
Large Scale Studies at the Pilot Plant - Large scale batch and contin-
uous flow experiments were conducted to verify results of the small scale
studies and to examine the other practical problems that may be encountered
with large scale installations. In the large scale batch studies,
ammonium chloride and poultry waste suspensions were used while in
continuous flow systems, only poultry wastes were used.
104
-------
40 -i
?30
u.
O
o:
UJ
x
u.
(O
I
§
10-
14,700 mg/l
8500 mg/l
10,400 mg/l
TOTAL
SOLIDS CONTENT
AS NOTED
POULTRY MANURE SUSPENSIONS
.2
.4 .6
KD/ HOUR
.8
FIGURE 40
SOLIDS CONTENT OF A LIQUID AS IT
AFFECTS K
105
-------
TABLE 10
CHANGE IN WASTE CHARACTERISTICS AS A RESULT OF AMMONIA DESORPTION
(percent)
LIQUID
Poultry manure
wastewater
TOTAL
SOLIDS
+38 to -22
COD
12 to 80
TOTAL
NITROGEN
AMMONIA
NITROGEN
14 to 58 39 to 77
ORTHO-
PHOSPHATE
5 to 80
Dairy manure
wastewater
+6 to -19
25 to 45
6 to 39 19 to 67
5 to 34
Oxidation ditch
mixed liquor
treating
poultry wastes
+53 to -6
7 to 73
9 to 49
9 to 68
12 to 60
decrease is noted by a minus or no sign, increase by a plus sign.
NOTE: This data was obtained from batch study experiments at temperatures from 10°C to 35°C.
was made to control the pH. Air flow rates varied as noted in Figure 36 and Table 9.
No attempt
-------
In these experiments, suspensions of poultry waste with solids concen-
trations ranging from 20 to 120 grams/liter were used. The compressed
air systems at the pilot plant were used for aeration and the rates of
air flow used ranged from 3-30 SCFM. A steel tank with a capacity of
500 gal. was used as the reaction vessel in both the batch and continuous
flow studies. The volumes of liquid used ranged from 150-450 gallons so
that various air flow rates to liquid volumes could be evaluated. The
experiments were conducted at ambient laboratory temperatures, 18-22°C.
In the continuous flow studies, the rates of air flow varied from 0.4-12
SCFH/gallon with detention times varying from 25-120 hours. Sodium
hydroxide and calcium hydroxide were used to adjust the pH of the liquids
and the range of initial pH values used in the studies ranged from 8.5
to 12.0. In the experiments, pH control was not maintained, and these
pH values decreased naturally in the runs as the ammonia was desorbed.
a) Batch Units - Both fresh and "aged" poultry manures were used in
these experiments. When fresh poultry waste was used, the ammonia con-
tent of the suspension increased (Figure 41). This was due to the con-
version of organic nitrogen to ammonia nitrogen. In view of this, it
was difficult to determine the value of 1C, of suspensions of fresh
manures. When mercuric chloride, an inhibitor, was added to the liquid,
it prevented the biological conversion of organic nitrogen and facili-
tated the measurement of 1C.
To overcome this problem, "aged" manures were used in subsequent experi-
ments. The "aged" manure was prepared in the following manner. Enough
water was added to fill the voids of varying amounts (25-125 Ibs) of
fresh poultry waste. This mixture was held approximately ten days to
permit maximum conversion of organic nitrogen to ammonia. This "aged"
mixture was suspended in 150 gallons of water in a tank and desorption
studies were made. Lime and sodium hydroxide were used in dif-
ferent experiments to control pH. When the quantity of the poultry
manure exceeded 50 Ibs per 150 gallons, the viscosity of the suspensions
increased after about 24 hours of aeration coincidental with a decrease
in the value of 1C (Figure 42). The change in viscosity occurred with
both bases and was due to the solubilization of organic matter.
The results of these large scale batch studies (Figure 43) confirmed
previous relationships indicated in Equation 44. The slopes of these
lines, the "b" value in Equation 44, ranged from -3.4 to -3.7 with an
average of -3.5. However, these values may differ if problems of repre-
sentative sampling are encountered.
When the supernatant liquid, containing almost all the ammonia nitrogen
of the aged poultry manure was decanted and used for stripping, the prob-
lem of thickening of the liquid was overcome. It was also found that
the value of 1C, remained the same even after five days of aeration. The
viscosity of the liquid did not change during these runs.
107
-------
420 -i
FRESH POULTRY MANURE
LIQUID VOLUME - 1450 liters
AIR FLOW RATE - 7 SCFM
TOTAL SOLIDS - 62,000 mg/1
LARGE SCALE BATCH STUDY
CONTROLLED pH
320
20 30
HOURS OF DESORPTION
40
50
6<
FIGURE 41
AMMONIA VARIATIONS DURING THE
DESORPTION OF FRESH POULTRY MANURE WASTEWATER
108
-------
1000-
800-
| 600 -I
z
E
i 400]
O
5
20O
CONTROLLED pH
AGED POULTRY MANURE
AIR FLOW RATE - 7 SCFM
LIQUID VOLUME - 1450 Liters
LARGE SCALE BATCH STUDY
DUE T0
Ul
O
O
O
VISCOSITY CHANGE
100
10 15 20 25
PERIOD OF AERATION - hrs.
4
FIGURE 42
CHANGE IN THE AMMONIA DESORPTION RATE
DUE TO CHANGES IN VISCOSITY
30
109
-------
.0451
.030
.020
-010
.008
.006
,004
.002
.001
27 SCFM / I50gal.
AGED POULTRY MANURE
CONTROLLED pH
LARGE SCALE BATCH STUDIES
5 10 20
VISCOSITY - centipoises
FIGURE 43
EFFECT OF AIR FLOW AND VISCOSITY
ON KD - LARGE SCALE STUDIES
30 40 50
110
-------
b) Continuous Flow Units - Aged manure from 50 pounds of fresh manure
was mixed daily with 150 gallons of water and allowed to stand for two
hours. The supernatant was pumped into a storage tank for use in these
experiments. This operation reduced the solids content of the liquor
and prevented the "thickening" of the liquid during storage and air
stripping. In these experiments, air flow rates from 10 to 30 SCFM
were used and the pH controlled to levels in the range of pH 9 to 11
with sodium hydroxide. A dilute liquid solution of sodium hydroxide
was continually added to the desorption unit to control the pH to desired
levels. Thus when the continuous system had reached equilibrium, the
liquid in the desorption tank had a constant pH and a constant total
ammonia concentration. Different pH levels were obtained by adjusting
the sodium hydroxide input. The addition of about 10 ml of lubricating
oil per day was sufficient to suppress foaming. If the inflow of liquid
and alkali was not permitted to splash, foaming was greatly reduced.
The results of these continuous flow experiments verified that Equation
38 could be used to predict values of 1C using actual data and therefore
the Equation can be used for purposes of large scale continuous ammonia
desorption systems. These results also confirmed that the data obtained
in both laboratory batch and continuous flow units we re comparable to that
obtained in the larger scale studies (Figure 44). The data used to
prepare Figure 44 were obtained as follows. The laboratory batch and
continuous flow experiments had developed equations and relationships
that could be used to predict 1C under a variety of process conditions
(Figure 36, Equation 44, and Figure 40). Based upon the actual process
conditions in the large scale pilot plant batch and continuous flow
experiments, the expected 1C values were determined using the equations
and relationships developed from the laboratory experiments. These
expected 1C values were compared to the 1C values obtained from the
pilot plant experiments in Figure 44.
"a
up
values from each experiment should be system specific, dependent
>on tank geometry, type of diffuser, and turbulence. The data obtained
in the laboratory and pilot plant experiments were obtained with different
diffusers, tank sizes and geometry, and turbulence levels. Any differ-
ences in KD due to diffuser type and tank size and geometry were not
observed in this study. This may be due to the gross nature of these
experiments and the fact that adequate turbulence and air:liquid contact
did occur. In such situations, the air quickly becomes saturated with
a very small amount of ammonia. The transfer of ammonia from the liquid
to the gas phase is essentially complete within the first few millimeters
of the ascent of the bubble in the liquid (34).
The relatively good correlation between desorption coefficients obtained
from laboratory and pilot plant studies suggests that data from labora-
tory batch scale units can be used, with the relationships developed in
this study, to estimate the desorption coefficients and hence ammonia
111
-------
oc
o
X
2
UJ
eo
CO
O
.04
Z
O
Q
UJ
CD
UJ
3
§
o
UJ
o
UJ
OL
45° RELATIONSHIP
.03-
.02-
.01
0
DIFFERENT DIFFUSERS,
TANK VOLUMES, AND
TURBULENCE WERE USED
IN THE LABORATORY AND
PILOT PLANT EXPERIMENTS
.01
.02
.03
.04
OBSERVED VALUE IN PILOT PLANT TRIALS - KQ PER HOUR
FIGURE 44
1 OF KD VAl
LABORATORY AND PILOT PLANT STUDIES
COMPARISON OF KQ VALUES FROM
112
-------
removals that will occur in large scale batch and continuous flow units.
Greater verification of such correlation would be desirable.
PREDICTIVE RELATIONSHIPS
General - Basic data on mass transfer properties of liquids are necessary
for either designing equipment suitable for desorption of ammonia from
wastewaters, or for assessing the magnitude of ammonia desorption in
treatment systems in which ammonia losses occur. Some information is
available in chemical engineering literature on absorption of ammonia
in liquids. Little work, if any, has been done on ammonia desorption
and data on rates of interchange of gases other than oxygen during
aeration are meager (41). In general, it is necessary to determine
desorption coefficients by direct experiment or to rely on semi-empirical
relationships.
The rate of desorption of ammonia from solutions is a function of varia-
bles which include (a) solubility and degree of dissociation of ammonia,
(b) temperature, (c) rate of aeration, (d) viscosity of the liquid and
(e) surface tension of the liquid. These studies on desorption of
ammonia from water and wastewaters by air stripping have provided useful
information on the important factors which govern the rate of removal
of ammonia by aeration. Using the relationships and equations developed
in this study it is possible to predict the ammonia removal that will
occur at different air flow rates, temperatures, pH levels and with
wastes of different characteristics. While it is true that system
geometry will affect the desorption rates, the different tank sizes
and shapes and types of diffusers did not appear to have a measurable
different effect on the desorption coefficients. For the systems used
in this study, the five parameters noted above had a greater effect than
did system geometry.
Fresh animal wastes have only a small part of their nitrogen in the form
of ammonia. Microbiological transformations occurring in the wastes
result in the conversion of a large part of the organic nitrogen to
ammonia. In an "aged" poultry waste, as much as 50 percent of the total
nitrogen could be in the form of ammonia. The mathematical models
developed in this study are based on the assumptions that the process of
conversion of organic nitrogen to ammonia is complete and that the
oxidation of ammonia to nitrites and nitrates is negligible. Other
assumptions implicit in the development of the model are that ammonia
acts as an ideal gas in these wastewaters, and that it obeys Henry's
law when corrections are made for the degree of ionization. Modifica-
tions of the model should be explored when these assumptions do not hold.
For the sake of brevity, the several factors affecting desorption of
ammonia are discussed separately.
EH, Temperature, and Air Flow - The resistance to mass transfer of ammonia
from a liquid phase to a gaseous phase results from the resistances of the
liquid and gas films at the interface. The gas film resistance increases
113
-------
with increasing solubility of the solute. Since ammonia is highly solu-
ble in water, the gas film resistance is very high compared to the liquid
film resistance. Even a very small amount of ammonia desorbed from the
liquid is sufficient to saturate the gaseous phase. Therefore, a large
volume of air must be passed through the liquid to remove substantial
amounts of ammonia (34, 42).
The kinetics of the mass-transfer process for desorption of ammonia from
waste or wastewater are complicated by ionic equilibria. In the reaction
of ammonia with water to form ammonium hydroxide, the quantity of ammonia
present in the undissociated form is dependent upon the pH value of the
solution and the dissociation constant for ammonium ion (Equation 14).
At a given pH value, the proportion of the undissociated ammonia increases
with increasing temperature.
The amount of ammonia removed per air flow used for desorption, dq/dV,
can be related to the initial total ammonia concentration, C, and the
fraction of undissociated ammonia, F, as follows
r • $ • V (45)
where K-, is the mass-transfer coefficient of this relationship. The term
~C~ ' dV" 1S ^e Percent Of *ne ammonia removed from a volume of liquid
by a volume of air. Equation 45 is independent of time and permits an
understanding of the quantity of air needed for the removal of ammonia.
The percent ammonia removed from a liter of poultry and dairy manure
wastewater by a cubic foot of air is noted in Figures 45 and 46. The
data in Figure 46 represent results from individual runs at the noted
temperatures. Figure 45 was obtained by pooling the results from many
runs conducted at the noted conditions. Variations within a run were
caused by variations in sampling and possible desorption while the
samples waited for analysis.
Based on theoretical relationships, the percent removal should reach a
plateau when all of the ammonia is in the undissociated form. This
occurs at pH values above 10.5 to 11 depending upon temperature. The
expected relationship was observed with specific runs of dairy wastewater
(Figure 46) but not with poultry waste (Figure 45), possibly due to a
masking of the relationship by pooling of the data. The maximum percent
of ammonia removed by a cubic foot of air increased with temperature
(Figures 45 and 46) and was about 1.0 for dairy manure wastewater and
poultry manure wastewater at a temperature of 20°C. The data noted in
Figures 45-47 resulted from batch studies using aged manure The pH
was not controlled.
114
-------
0 3.0
a
POULTRY
UJ
I-
o 2.C
a:
a
1.5-
l.O-
Q
U
1 «
01
to
I
z
8.5
35 °C
30°C
25°C
20°C
I5°C
IO°C -
- *
35°C
9.0
e '30°C
IO°C
—i—
9.5
10.0
10.5
11.0
—i
11.5
PH
FIGURE 45
AMMONIA REMOVED PER VOLUME OF AIR
AS AFFECTED BY pH AND TEMPERATURE -
POULTRY MANURE WASTEWATER
115
-------
2.0!
DAIRY
u.
O
El
3 5 '
>
03 a: i.O
o u
ui t
1« 5
UJ ^ -°
IT
9.5
35° C
IO.O
pH
10.5
FIGURE 46
AMMONIA REMOVED PER QUANTITY
OF AIR - DAIRY MANURE WASTEWATER
-------
The quantity of ammonia removed by a unit volume of air was directly
proportional to the total ammonia concentration in the liquid (Figure 47)
and was related to the temperature of the liquid.
The quantity of ammonia removed by a unit volume of air per fraction of
undissociated ammonia (r~ ' TJv F^ ' 1*>e-> ^he mass transfer coefficient
(Ky) of Equation 45, varied with respect to temperature in a linear
manner (Figure 48) over a broad temperature range. The data noted in
Figure 48 was obtained from a number of batch experiments with poultry
manure wastewater and over a wide pH range. The pH in these experiments
was uncontrolled. A similar linear relationship of KV with temperature
was obtained when the data from dairy manure wastewater was plotted.
Effect of Rate of Aeration at Different Temperatures - At any given
temperature, higher values of 1C. can be obtained by increasing the rate
of aeration. The relationship of 1C to air flow rate (A) per liter of
liquid (Lg) was linear (Figure 36a) and can be expressed as
where D is the intercept on the X axis and is numerically equal to the
rate of desorption of ammonia when there is no aeration. The value of
D is small compared to MIC when diffused aeration is used for desorption.
When this occurs, the equation can be rewritten as
A = MKn (47)
LQ u
The desorption rate, KD, increases with an increase in air flow rate
and decreases as the liquid volume of the unit increases. All parameters
in Equation 46 and 47 must be obtained at the same temperature.
The value of M can be determined from specific laboratory or larger
scale experiments with known A to LQ ratios and KD values. The value
of M depends upon temperature and decreases with increasing temperature.
Results of our observations (Table 11) indicate that the M values for
water, poultry manure wastewater, and dairy manure wastewater were
similar at common temperatures. All of these liquids had low solids
contents. The value of M for the oxidation ditch mixed liquor, which
contained a high solids content, was different from the values obtained
for the other liquids.
117
-------
CD
±20
u
X
0»
15
Q
LU
o
Ui
1 10
o
U.
o
5
O
0
FIGURE 47
QUANTITY OF AMMONIA REMOVED PER
QUANTITY OF AIR RELATED TO
TEMPERATURE AND AMMONIA CONCENTRATION
30°C -
25°C -
20°C -
I5°C -
pH 10-1O.5
ZO°C
I5°C
POULTRY WASTE WATER
IOO
200
300
400
CONCENTRATION OF AMMONIA
IN UNTREATED LIQUID
500 600
mg N/ liter
700
800
-------
.041
.03-
-[*• .02-
cri >
•o-o
— lo
.01-
10 15 20 25 30
TEMPERATURE - °C
35
FIGURE 48
THE VARIATION OF Ky (EQUATION 45)
WITH TEMPERATURE
119
-------
TABLE 11
VALUES OF M FOR THE EQUATION
A/LQ = M
LIQUID
Water plus
ammonium chloride
Poultry manure
wastewater
Dairy manure
wastewater
Mixed liquor from
an oxidation ditch
treating poultry waste
M (SCF/L)
1Q°C 15°C 20°C 25°C
77.2 63.0 46.8
97.8 72.7 67.9 49.3
72.6 68.9 38.5
.0001 49.2
30°C 35°C
29.2 27.5
23.5
LIQUID
Nater plus
ammonium chloride
Poultry manure
wastewater
Dairy manure
wastewater
Mixed liquor from
an oxidation ditch
treating poultry waste
TOTAL SOLIDS CONTENT
DURING THE DESORPTION EXPERIMENTS
(mg/1)
4000 - 8000
8000 - 10,000
28,100 @ 15°C
19,000 @ 20°C
120
-------
The data in Table 11 resulted from pooled data of batch studies in
which the pH was not controlled. The air to liquid ratios used in
these studies ranged from 3 to 36 SCFH of air per liter of liquid. The
values of M were obtained by fitting the data points to a straight line
using the least squares method.
For specific air to liquid ratios, the M value and therefore the 1C
value appears dependent upon the solids content of the liquids rather
than on the type of liquid. Using the M values determined in these
studies, it may be possible to predict 1C for any waste having a solids
content similar to those contained in the liquids that were used. The
values of M may be system dependent.
When the values of M, i.e., the slope of the IC-air flow to liquid
volume lines (Figure 36a), are plotted against temperature, a straight
line was obtained (Figure 44) which can be expressed as
MQ = me + p (48)
D
MQ is the slope of the air flow/liquid volume to 1C relationship at a
temperature e and m and p are constants. Values of these constants for
our experiments are presented in Table 12. Equation 48 permits the
determination of M and hence 1C at any temperature and expands the use-
fulness of Equation 46 to permit the evaluation of the effect of different
air flow rates and liquid volumes over a wide temperature range. The
differences in "m" noted in Figure 49 are not thought significant. It
is likely that the average value can be used for water, poultry manure
wastewater, and dairy manure wastewater.
The value of 1C is related to both temperature and the rate of air flow.
The ratios of the values of 1C at different temperatures to the value of
Kg at 20°C are presented in Table 13. The results indicate that indi-
vidual values of 1C are increased by a factor of 1.5 to 2 when the temp-
erature is raised by 10°C. This relationship suggests that the following
empirical relationship may be employed for temperature and 1C in the
temperature range of 10-35°C.
Kn = Kn E(62"9l) (49)
D
•*•>,
where K and K are the desortion rates at the same rate of air flow
Kn and Kn are the desorption
ul D2
at temperatures QI and 62. The values of E from these experiments are
noted in Table 14. The values of E are in the range of 1.06 to 1.065
121
-------
1001
m- SCF/ L/°C
20
10
POULTRY
WASTE
20 30
TEMPERATURE - °C
FIGURE 49
SLOPE OF THE KD - AIR FLOW RATIO
AS A FUNCTION OF TEMPERATURE
TABLE 12
VALUES OF m AND @ FOR THE
EQUATION M = me + g
LIQUID
Water plus ammonium chloride
Poultry manure wastewater
Dairy manure wastewater
m
-3.04
-2.86
-2.65
123.1
121.8
114.0
122
-------
TABLE 13
RATIOS OF KD AT NOTED TEMPERATURES TO
KD AT 20°C FOR SPECIFIC RATES OF AIR FLOW
(mean values)
AIRATE°W TEMPERATURES - °C
(SCFH/L)
6
12
20
10
38
40
47
15
.77
.70
.70
20
1.0
1.0
1.0
25
1.54
1.40
1.40
30
1.73
1.60
1.74
35
1.92
2.08
1.91
TABLE 14
VALUES OF THE TEMPERATURE COEFFICIENT E
IN THE EQUATION 1C = Kn E^2"6^
U2 U1
AIR FLOW
(SCFH/L) E_
6 1.064
12 1.065
20 1-060
123
-------
with a mean of 1.063. This temperature factor, E, for desorption is in
the same range as constants obtained for changes in biological reaction
rates (43-45).
The relationship between Kn and temperature at a specific air flow rate
(Figure 36b) appears to be a straight line. However, both a logrithmic
and straight line can be used to obtain a good fit of the data in the
10-35°C temperature range studied. Both types of relationships appear
able to be used with equal facility for estimating the effect of temp-
erature on KD in ammonia desorption.
The temperature-Kn plots (Figure 36b) have a positive intercept on the
temperature axis. This suggests that it will not be possible to desorb
ammonia at temperatures below about 5°C. It is possible to develop an
empirical relationship to relate K, air flow rate and temperature. The
relationship would be for data collected on air flow rates from 6-20
SCFH/liter and temperatures from 5-35°C. For practical purposes, the
following empirical equation can be used to relate KD, air flow rate,
liquid volume and temperature:
[0.091 p-+ 0.062(6-5)]
= 0.021e Q (50)
The units of the parameters in Equation 50 are: A - standard cu. ft. of
air flow per hour, Lg - liquid volume in liters, 0 - temperature in °C,
and KD - per hour. This Equation represents data obtained from the lab-
oratory and pilot plant batch and continuous experiments described pre-
viously and may not be representative of other conditions or wastes.
Table 15 compares the results obtained with this equation to the results
obtained from actual experiments in this study. The comparison is good
indicating that Equation 50 can be used for predictive purposes.
The intercepts on the x axis of KD-air flow lines (Figure 36a) show that
desorption of ammonia can take place above the minimum temperatures even
if no air is diffused through the liquids. When the air flow is zero,
KD also is zero and the only loss of ammonia occurs due to diffusion
through the surface of liquid.
Results of brief experiments to estimate the ammonia loss under quiescent
conditions have been presented in Figure 37. Data from these experiments
indicated a DQ value of 0.003 per hour per sq. cm. at 25°C. This corre-
sponds to a diffusivity value of 7.5 x 10"4 ft2 per sec. which is about
three times values reported for ammonia in chemical engineering references
When air flow is zero, Equation 50 reduces to
124
-------
TABLE 15
COMPARISON OF EXPERIMENTAL AND PREDICTED KQ VALUES
AIR FLOW RATE TEMPERATURE KD ^per hour^
(SCFH/liter) (°C) Predicted* Experimental**
20
20
20
13.3
20
20
20
7.3
12
6
6
16
16
13.3
4
18
21.5
20.5
19.0
26
31
36
20
20
20
28
20
26
21
20
0.290
0.361
0.339
0.168
0.477
0.650
0.886
0.103
0.159
0.092
0.151
0.228
0.331
0.190
0.077
0.284
0.378
0.354
0.228
0.473
0.594
0.834
0.067
0.179
0.085
0.119
0.221
0.325
0.219
0.082
Predicted by using Equation (50)
*
The average value of KD observed in experiments
125
-------
(51)
in which KQ now represents an estimate of DQ.
Surface Tension and Viscosity - Physical properties, like surface tension
and viscosity influence mass-transfer coefficients (46). For a mole-
cule of dissolved gas to escape from the liquid phase into the zone of
saturated vapor, it must have the necessary energy to overcome the
barrier of surface forces. By increasing the temperature it is possible
to decrease this surface tension. While this may partly explain higher
rates of desorption observed at elevated temperatures, no direct corre-
lation between surface tension and desorption rates was found in this
study.
Viscosity was found to affect the rate of desorption. Raising the
temperature of a liquid lowers its viscosity and favors more intimate
contact between the two phases. Straight and parallel lines were obtained
by logarithmic plots of viscosity and 1C at different air flow rates
(Figures 39 and 43). Similar results have been obtained in a study of
factors governing the rate of oxygen dissolution in viscous liquids (47).
The theoretical interpretation of these relationships remains unknown.
However, this empirical relationship provides a tool to predict the rate
of desorption at different air flow rates and viscosities of liquids.
The intercepts on ordinates of the viscosity-IC curves are the logarithm
of the values of KQ when viscosity is 1 centipoise. The slopes of all
the curves generally had a value of -3.5 (Table 9). This high value
suggests that, in addition to the viscosity, other factors such as
density, surface tension and diffusivity, which also change as viscosity
varies in a liquid, may have had an effect on 1C.
By knowing the viscosities of the suspensions of wastes, and using
Equation 44, it is possible to predict the value of 1C of a liquid (1C )
if the KD of a similar liquid (KD ) and the viscosities of both liquids
(PI and y2)are known.
I/
D v-i -b
<52>
This relationship was developed for air-tap water, air-poultry waste,
and air-dairy manure wastewater suspensions. Equation 52 is valid un
126
-------
the experimental conditions described in this study and only when the
air flows to be used for the desorption of the two liquids are the same.
Equation 52 permits extrapolation of available data. Desorption coef-
ficients obtained with one waste can be used to estimate desorption
coefficients of different wastes or of wastes modified by in-plant
changes. The equation also permits better use of laboratory time in
obtaining 1C values since the equation can be used to relate data from
different wastes and only a small amount of confirmatory data may be
needed.
In our studies, it has been observed that unless the viscosity differed
by more than 20% during a run, no discernible difference in 1C was
apparent. The value of "b" was found to be 3.5 over a range of 10.8
to 76 SCFH/gal. The value of "b" was lower than 3.5 when the air flow
rate was less than 10.8 SCFH/gal.
SIGNIFICANCE OF THE RESEARCH
Combined Predictive Relationships - The equations described in this
section coupled with those developed earlier in the report can be
utilized to obtain general predictive relationships for the desorption
of ammonia from wastes. The predictive relationships can be used to
estimate the magnitude of the design parameters to remove ammonia from
a specific waste. It should be recalled that desorption coefficients
are a function of the aeration system that is used. Care should be
used in extrapolating these relationships beyond their intended use.
a) General Equations for 1C - The desorption coefficient, KD, represents
the rate at which ammonia is lost from an aerated system. The actual
values of KQ can be varied by altering environmental factors such as
temperature, viscosity, air flow, and type of aeration system. Know-
ledge of the value of Kp for a given liquid permits the determination
of the time required to remove a given quantity of ammonia (Equations
27 and 29) in either batch or continuous flow systems. The ability to
predict 1C is essential to the practical use of ammonia desorption
equations and systems.
If ammonia desorption is contemplated for an available waste, laboratory
experiments, such as those described in this research, can be used to
obtain values of 1C over a variety of contemplated air flow rates and
liquid volumes. Experiments in which the pH is not controlled can pro-
vide information on Kp if Equation 38 is used. If the pH is controlled,
then either Equation 27 or 29 can be used to determine KD depending upon
whether batch or continuous experiments are used.
The results of this research have demonstrated (Figure 44) that Kg values
obtained from laboratory batch scale experiments can be used with confidence
127
-------
in both large scale batch and continuous flow experiments. The relatively
good correlation between KQ values obtained from laboratory and pilot
plant experiments suggests that data from laboratory batch scale units
can be used with the relationships developed in this study to estimate
desorption coefficients in large scale systems. This use eliminates
the need for a multitude of confirming experiments in large scale
diffused aeration ammonia desorption systems. The KD values obtained
from laboratory units using diffused aeration for desorption can be used
as close estimates of the values that will occur in larger systems for
predictive purposes.
The 1C values obtained at one temperature can be extrapolated to values
at other temperatures using Equation 49. The effect of both temperature
and air flow rates on KD can be estimated by using Equation 50. Undoubt-
edly one would like to conduct confirmatory experiments at other temp-
eratures, air flow rates, with systems of other sizes, and with wastes
of different characteristics. However, the closeness of the results
obtained in this study indicate that such confirmatory experiments may
be held to a minimum.
If the physical characteristics of a waste, such as viscosity, may change
due to different process or waste changes or if it is desired to estimate
KD values for wastes of other characteristics, Equation 52 can be used.
Only a measurement of viscosity is needed for an estimate of the KD for
the second waste. It is assumed that the KQ and viscosity of the first
liquid are known and that the same air flow rates and liquid volumes
are to be used. With Kn known, the effect of different temperature,
?
air flow rates, and liquid volumes can be mathematically evaluated using
other equations to obtain the optimum 1C for the desired ammonia removals.
b) Relationships to Predict Ammonia Loss - With 1C known or estimated
by the above approaches either Equation 27 or 29 can be used to deter-
mine the change in total ammonia concentration that will occur in a
given aeration time. The pH of the solution is another decision
variable that is included in the use of these equations. In this manner,
the most appropriate design relationships to accomplish a necessary
degree of ammonia removal can be estimated.
c) Application to Practice - In practice, the amount of ammonia that
must be removed is established by environmental quality considerations
such as effluent or stream quality criteria for discharge to surface
waters or the amount that can be included in land disposal of wastes
without being excessive. With the needed removal known, i.e., C, and
C2, the engineer has a number of decision variables to consider in the
design of an ammonia desorption system. These variables include air flow,
128
-------
pH, temperature, and time of desorption. For a specific waste, the flow
and physical characteristics such as viscosity, surface tension, and
solids content should vary only within a reasonable range and in addi-
tion are usually unable to be controlled or altered by the design
engineer. The engineer has the challenge of determining the best set
of design variables to meet the needed removal at minimum cost.
The steps that would be taken in arriving at the best design are:
i) using samples of the waste, determine 1C under possible air flow
and liquid volume relationships; diffused aeration laboratory experiments
may be satisfactory for this purpose.
ii) investigate other possible values of K over the temperature range
likely to be found in practice (Equation 49), especially to determine
the effectiveness of removal at low temperatures and to estimate the
potential of increasing the temperature of the waste prior to desorption.
iii) use either Equations 46 and 48 or Equation 50 to investigate other
possibilities to vary air flow and liquid volume to obtain optimum
desorption rates.
iv) with possible 1C values known, investigate the pH and time rela-
tionships that will obtain the desired ammonia removal in batch systems
(Equation 29).
It should be noted that if the maximum possible ammonia removal is desired,
then both 1C and F should be as large as possible, i.e., maximum air flow
rate, temperature, and pH. However, if less than maximum removals are
adequate, then there are a number of possible design trade offs between
1C, F, and time to obtain these removals. The decisions would be made
on the basis of the relative costs of aeration equipment, base to adjust
the pH, tankage to obtain the necessary time, and heat to adjust the
temperature. It is important to note that maximum pH levels, aeration
rates, time, or temperature may not be necessary if intermediate ammonia
removals are adequate. Such is likely to be the case when ammonia
removal from agricultural wastes is practiced since the land is the
disposal point for most of these wastes. Under proper soil and crop
management practices, maximum removal of nitrogen prior to disposal is
not necessary. Conservatively the nitrogen applied to soil in wastes
should be no greater than the amount that will be removed in the crops
to be grown in the area. If amounts above this level are added, the
potential for nitrogen leaching to the ground waters is enhanced.
v) scale up the best data obtained in i-iv above for a system to meet
the required ammonia removal needs.
vi) determine the capital and operating costs of the various alternatives
129
-------
and determine the optimum combinations of variables to produce the
least expensive systems.
An example of the types of alternatives that could be involved in the
removal of ammonia from poultry manure wastewaters can be accomplished
by using data obtained in this study. Table 16 was prepared to illus-
trate the patterns that could result from different combinations of
air flow rate per unit of liquid, pH, temperature, and percent removal.
Thus there are many decisions that can be made to obtain a specific
ammonia removal, each requiring different desorption times to accom-
plish the desired results. A portrayal of some of the relationships
is presented in Figure 50.
The data in Figure 50 and Table 16 were developed using the desorption
coefficients and predictive relationships developed in this study. A
batch system treating dilute poultry waste suspensions is assumed. The
Table is presented in terms of percent ammonia removal to make it useful
for wastes of different ammonia concentrations.
Little difference in desorption time occurred at pH values of 10-11
over temperatures from 10° to 25°C. Below a pH of 10, the time to
accomplish a specific removal increased rapidly. Higher removal effi-
ciencies required increased desorption times. The time to accomplish
75% removal was double that to accomplish 50% removal at 20°C. In
the same manner, it would take twice as much time to accomplish 99%
removal as to accomplish 90%.
A change in temperature affected the desorption time. At a pH of 10 to
11, a 10°C temperature drop doubled the time necessary to obtain a
specific removal. At a pH of 8-9, a 5°C temperature decrease doubled
the desorption time to achieve the desired removal. The larger tempera-
ture effect at the lower pH values occurs due to the cumulative effect
of pH on the free ammonia concentration and the effect of temperature
on
Decreasing the air flow rate by 5 SCFH/L increased the desorption time
by about a factor of 1.6 while decreasing the air flow rate by 10
SCFH/L increased the time by about a factor of 2.5. The changes noted
in Table 16 follow theoretical relationships described in the earlier
sections.
The data in Table 16 can be used to observe the efficacy of ammonia
desorption for typical poultry manure wastewater. This manure waste-
water would have about 4500 mg/1 of total solids, about 600 rng/1 of
NH4-N, and about 1200 mg/1 of total nitrogen (organic plus ammonia).
If the wastewaters were "aged" most of the ammonia production had occurred,
The remaining organic nitrogen could be considered as not readily
decomposable.
130
-------
40CH
50% REMOVAL
10 SCFH
pH 9
10 15 20 25
TEMPERATURE -°C
. 99 %
20°C
15 SCFH
o
UJ
H _, 300-
o z
•z. o
,_ t-
J- Q.
OT K
-J O
a. en
o o 20° '
O i
0 <
0 0
en 2E
K < 100-
o
X
r>-
\
\
\
\
\ \
\ L—- 90 %
r\
i \
\ \
f \^\ 70 %
\^\ \
\ \ \
\ \ V- 50%
\NvX\
^ \^ x.
8
10
pH
FIGURE 50
DESORPTION TIME TO OBTAIN SPECIFIC
AMMONIA REMOVALS AS A FUNCTION OF
TEMPERATURE AND pH
-------
TABLE 16
TIME REQUIRED TO OBTAIN SPECIFIC
AMMONIA REMOVALS AS RELATED TO ^
pH, AIR FLOW RATE, AND TEMPERATURE
AIR FLOW RATE TIME REQUIRED - HOURS
(SCFH/liter of
liquid)
5
10
15
20
5
10
15
20
5
10
15
20
5
10
15
20
*
Based on ammonia desorption constants developed in this study assuming
ammonia is to be removed from a dilute poultry manure suspension in a
batch desorption unit. This Table was prepared to illustrate the relative
effect of the variables noted.
132
PH
8.0 8.5
75% removal
446 150
284 96
180 61
114 38
80% removal
517 174
330 110
210 70
132 44
90% removal
740 250
471 159
300 100
190 64
99% removal
1480 500
940 317
600 200
380 1 28
9.0
at 20°C
59
38
24
15
at 20°C
68
43
27
17
at 20°C
98
62
40
25
at 20°C
196
125
79
50
9.5
30
19
12
7.7
34
22
14
8.9
50
32
20
12
99
63
40
25
10.0
21
13
8r4
5.3
24
15
9.7
6.2
34
22
14
8.8
69
44
28
17
10.5
18
11
7.2
4.6
20
13
8.4
5.3
30
19
12
7.6
59
38
24
15
11.0
17
11
6.8
4.3
20
12
7.9
5.0
28
18
11.4
7.2
55
36
23
14
-------
TABLE 16 continued.
AIR FLOW RATE TIME REQUIRED - HOURS
(SCFH/liter of
liquid)
5
10
15
20
5
10
15
20
5
10
15
20
5
10
15
20
5
10
15
20
PH
8.0 8.5
50% removal
770 252
488 160
309 101
196 64
50% removal
400 136
255 87
162 55
103 35
50% removal
223 75
142 48
90 30
57 19
50% removal
117 41
74 26
47 17
30 10
50% removal
59 22
37 14
24 8.
15 5.
9.0
at 10°C
90
57
36
23
at 15°C
50
32
20
13
at 20°C
30
19
12
7.5
at 25°C
17
11
6.9
4.4
at 30°C
9.8
6.2
7 4.0
5 2.5
9.5
39
25
16
10
24
15
9.6
6.1
15
9.5
6.0
3.8
9.6
6.1
3.9
2.4
6.1
3.9
2.5
1.6
10.0
23
14
9.2
5.8
15
9.7
6.1
3.9
10
6.6
4.2
2.7
7.2
4.6
2.9
1.8
5.0
3.2
2.0
1.3
10.5
1
i
1
18
11
7.1
4.5
12
7.9
5.0
3.2
8.9
5.7
3.6
2.3
6.4
4.1
2.6
1.6
4.6
2.9
1.9
1.2
11.0
16
10
6.5
4.1
11
7.4
4.7
3.0
8.5
5.4
3.4
2.2
6.2
3.9
2.5
1.6
4.5
2.9
1.8
1.1
133
-------
Desorption of 50% of the ammonia would leave about 900 mg/1 total nitro-
gen while desorption of 90% would leave about 660 mg/1 of the total
nitroqen. Even under maximum desorption conditions, one-half or more
of the initial total nitrogen will remain. While a considerable amount
of nitrogen has been removed from the waste, the land still is the most
logical disposal point for these treated wastes. The remaining nitrogen
conversion and utilization will take place in the soil.
Ammonia desorption can reduce the amount of nitrogen in wastes spread
or disposed of on the land thereby decreasing possibilities of
excess nitrogen and environmental quality problems. In addition, the
knowledge obtained in this study concerning ammonia desorption can be
useful in many other studies.
d) Other Implications - The relationships obtained in this study have
provided auxiliary but important information on the performance of
systems in which ammonia is removed, either intentionally or uninten-
tionally. In many natural systems, ammonia is lost as a result of other
processes. Examples would be losses from aerated biological treatment
systems, swift streams, impoundments, aerated odor control systems, and
waste storage units.
Equations 35 and 38 provide the opportunity to predict the pH changes
that will occur in a system as the ammonia is lost. This prediction
can be made for both aerated and quiescent systems since the appro-
priate KD values are available from this research.
Nitrogen balances now can be better evaluated since the equations in
this report permit better estimates of the ammonia lost from aerated
systems, flowing streams, or impoundments by desorption. With the
amount of ammonia volatilized known, one can better estimate its effect
on the local environment.
The effect of ambient temperature changes on current and future ammonia
desorption systems can be better determined. As a result the feasibility
of ammonia desorption systems in all parts of the country can be better
estimated.
Place of Ammonia Desorption as a Method of Treatment - If treated animal
wastes are to be discharged to surface waters, a high degree of removal
of organic matter, color, and nutrients will be required. Even with a
high percentage of removal of BOD, solids, and nutrients, considerable
amounts of contaminants may still remain in the treated animal wastes.
A more reasonable method for disposal of these wastes would be to dis-
charge them on land rather than into surface waters. The nutrients
still contained in the treated wastes can be incorporated into crop
growth. However, the basic requirements that have to be met for disposing
these treated wastes on land will be that the materials have character-
istics that would prevent excessive nitrogen in any runoff or in any
percolate to the ground water. The nutrients in wastes disposed of on
134
-------
the land should be roughly the amount that can be removed in any crops
grown on the land. This may require constraints on either the total
amount of wastes applied to the land or on the amount of nitrogen in
any wastes disposed of on the land. Denitrification of oxidized
nitrogen in the soil is another possibility to minimize nitrate loss
to the surface or ground waters. However, little is known about how
to manage this process in the soil. Therefore at this time, reliance
must be placed on methods to manage the nitrogen in wastes either before
or as wastes are disposed of on the land.
The ability to vary the nitrogen content of animal wastewaters, as out-
lined in this section of the report, will increase the alternatives that
are available for land disposal. An additional advantage inherent in
the process of ammonia desorption from wastes is the opportunity to pre-
cipitate soluble phosphates if lime or magnesia is employed to raise
the pH.
The quantities of wastes that are to be treated depend upon the nature
of a given livestock operation. Waste volumes of 3 to 28 gallons per
day per animal have been reported for cowsheds, dairies and milking
parlors, and 10 to 30 gallons per day per animal for swine farrowing
houses. The quantities of wastes produced by poultry are even less.
On an average about 0.05 and 5 gallons of waste/animal/day are produced
in poultry and swine operation respectively. The volumes of animal
wastes that will be produced will not be as large as municipal wastes.
The quantities of ammonia nitrogen present in these waters are con-
siderably more than in municipal wastes. If these wastes are not kept
under aerobic conditions putrefactive changes occur in these materials,
resulting in the production of offensive odors. Aerobic conditions
can control odor production.
If ammonia removal and aerobic treatment of animal wastewaters can be
combined, three purposes may be served: (a) removal of nitrogen from
the wastes, (b) odor control, and (c) a certain degree of biological
waste treatment. In combining ammonia removal and aerobic treatments,
diffused or mechanical aeration should be considered as a means for
removal of nitrogen from animal wastewaters.
The quantity of air required to accomplish a specific ammonia removal
is of interest when considering desorption of ammonia. The experimental
data (Table 16) collected in our laboratory on the air requirements for
desorption of ammonia from dilute suspensions of agricultural wastes
are useful in this respect. Some investigators (24-26, 48-51) examined
the process of ammonia removal for high ammonia removals, using aeration
tower for stripping. Their results are included in Table 17. The
quantities of air required for achieving about 90 percent removal have
ranged from 300 to 750 cu. ft. of air per gallon of the waste liquid.
The data in Table 17 has been obtained from available reports and from
this study. To provide a basis for comparison, data in the pH range
135
-------
TABLE 17
RELATIVE EFFICIENCIES OF AMMONIA DESORPTION
U)
CTl
Type of Liquid
OTHER INVESTIGATIONS
Wastewater from
petroleum industry
(48)
Secondary sewage
treatment effluent
(26)
Anaerobic digester
supernatant (51)
THIS STUDY -
DIFFUSED AERATION
Poultry waste
Dairy waste
Oxidation ditch
mixed liquor
Water
AERATION TOWERS
Poultry wastewater
(Run 25)
PH
10.5
9.4
>9.0
8.9
8.8
10.8 to 11
10.8 to 11
11.0
11.0
11.2
10.6 to 10
10.6 to 10
10.5 to 10
10.3 to 10
9.3
11.2
Temperature
(°C)
.0
.0
23
23
.4 20
.4 20
.0 20
20
21
21
Quantity of
air used
(ft3/gal. )
300
300
480
480
480
380 to 3040
750
535
430
437
77.5
77.5
233
77.5
279
279
Initial NH4-N
concentration
(mg/1)
100
100
100
100
100
25
24.5
31
28.9
850
703
321
150
560
460
440
Quantity of
NH. removed
(mgN/gallon)
329
132
>368
353
225
70
93.4
75.2
100.7
2685
682
256
195
510
73
1320
-------
TABLE 17 continued.
RELATIVE EFFICIENCIES OF AMMONIA DESORPTION
Type of Liquid
Process Efficiency
Ammonia mgN removed
removed per liter per
(%) cu ft of air
KvF
OTHER INVESTIGATIONS
Wastewater from
petroleum industry
(48)
co
""* Secondary sewage
treatment effluent
(26)
Anaerobic digester
supernatant (51 )
THIS STUDY -
DIFFUSED AERATION
Poultry waste
Dairy Waste
Oxidation ditch
mixed liquor
Water
AERATION TOWERS
Poultry wastewater
(Run 25)
85
34
>95
91
58
72
98
63
90
82
25
21
33
23
45
77
1.1
0.44
>0.76
0.76
0.47
0.18 to 0.023
0.124
0.142
0.234
1.586
8.8
3.3
1.4
6.6
2.9
4.7
0.0110
0.0044
0.0077
0.0073
0.0047
0.0073 to 0.0009
0.0048
0.0046
0.0081
0.0073
0.0125
0.0103
0.0096
0.0118
0.0063
0.0009
0.0119
0.0089
0.0274 to 0.0083
0.0309
0.0238
0.00754 to 0.00093
0.00496
0.00470
0.0083
0.0075
0.0135
0.0112
0.0112
0.0141
0.0139
0.00091
-------
from 10-11 and at the temperature of 20-23°C has been used. Some data
obtained at lower pH values have been included. Process efficiencies
have been variable with from 32 to 98% ammonia removals being obtained.
In general the highest quantity of ammonia removed resulted from wastes
that had the highest ammonia concentration in the untreated waste.
The atrmonia nitrogen content of these wastes generally was below 30 mg
per liter. Only in one of the studies was a waste used having a higher
ammonium content (about 100 mg/1). In this study, though lower removal
efficiencies were observed, the quantities of ammonia removed from a
gallon of waste, in terms of the air requirements, were significantly
higher. The main reason for the lower percent removals could be that
the agricultural wastes contained higher ammonia nitrogen contents.
In an effort to compare all the studies on a common basis, the amount
of ammonia removed per unit air flow per initial ammonia concentration
was used (Equation 45) thus
r ' • V <45>
This Equation is independent of time. These unit efficiencies are
expressed as K F in Table 17. To compare the different studies inde-
pendently of F, the values of K are also noted.
The KyF values for each study vary in each study and among the studies.
Generally the KyF values ranged between 0.004 and 0.012. The K values
also exhibited a similar wide variation with values ranging from 0.004
to 0.03. The KyF and KV values obtained in this study are within the
range of values found in other studies. The data in Table 17 indicates
that although the process efficiencies in each study varied over a wide
range, the unit efficiencies, Ky, were comparable. In general, higher
KV values occurred with high strength wastes and at pH values 10.2 or
less. Greater amounts of nitrogen removed per volume of air occurred
with wastes having a high ammonia concentration indicating that the
efficiency of air usage for ammonia removal was greater with wastes
having large ammonia concentrations.
Animal wastewaters are held in holding tanks for long periods of time
prior to disposal. Unlike ammonia removal from municipal wastes, short
detention times are not an important factor when dealing with animal
wastes. These long periods permit using lower air flow rates to accom-
plish reasonable ammonia removals since the ammonia removal per unit
volume of air was greater with wastes having high ammonia concentrations.
Even at low rates of air flow, it is possible to achieve a greater
removal of ammonia by raising the pH of the liquid. The optimal require-
ments of alkali have already been indicated. It is important to note
138
-------
that if lime or magnesium hydroxide is used to adjust the pH value of
the liquids, the quantities of resultant sludge can be large. While
higher pH values may afford ease of removal of ammonia, the disposal
of the resultant solids may create another problem. Although greater
removals of ammonia can be obtained by raising the temperature of the
liquid, it is doubtful whether it is practicable to heat animal wastes.
The data collected in our laboratory on the different factors affecting
removal of ammonia have provided useful information and predictive
equations which can be used to evaluate existing facilities for their
efficiency of ammonia desorption, and to aid the design of equipment
suitable for ammonia desorption.
139
-------
NITROGEN REMOVAL BY NITRIFICATION-DENITRIFICATION
INTRODUCTION
An estimate of the nitrogen and phosphorus contribution from various
sources indicates that domestic and industrial wastes are not the leading
sources of nitrogen and phosphorus in the environment (52). The nitrogen
content of the waste from chickens including broilers in 1970 is esti-
mated to be about 9 million pounds per day. Only a small portion of this
should reach surface and ground waters. There have been no estimates on
the amount of nitrogen reaching the surface or ground waters from poultry
wastes. With greater emphasis being placed in nutrient removal from
municipal and industrial wastewaters, nutrient control for agricultural
wastes may require investigation. At the present time, only a few
studies are being made on this problem. One large scale study is
underway at the San Joaquin Valley in California where biological
denitrification and growth of algae are being investigated.
Although nitrification followed by denitrification has been success-
fully demonstrated as a means of elimination of nitrogen from municipal
wastes, and it is recognized that this process may be of application in
the control of nitrogen in animal wastes, information on the process and
the parameters that control the process is relatively sparse. The
current study was undertaken to study the feasibility of nitrifying
and then denitrifying animal waste to minimize the nitrogen content
before it is eventually disposed of on land.
METHODS OF REMOVAL
General - Although there are several methods available for the removal
of nitrogen and for the treatment of animal wastes, none of the methods
are practiced to the fullest extent today. Some of the methods are
being tested on laboratory and pilot scale and the state of art of
removal of nutrients from animal wastes is in its infancy.
Traditionally the land has been the ultimate disposal medium for animal
wastes since they have value in maintaining and improving the soil tilth
and fertility. However, if uncontrolled spreading of manure is practiced,
the danger of contamination of surface waters via runoff from these
fields and contamination of ground waters by percolation is possible.
Three pounds of nitrogen and one pound of phosphorus were lost by runoff
from a ten ton per acre application of dairy manure on frozen soil
which had an 8 percent slope (53). In another study, manure applica-
tions on frozen ground resulted in losses up to 20 percent of the
nitrogen, 12 percent of the phosphorus, and 14 percent of the potassium
in the manure under conditions favoring maximum early runoff (54). One
of the ways to minimize nutrients in runoff is to remove nutrients
prior to spreading on the land.
A high degree of nutrient removal may not be necessary in animal waste-
water treatment facilities when the effluent is disposed of on land.
141
-------
In the soil a certain degree of nutrient removal takes place due to
synthesis of microbial and plant protoplasm. The uptake of nitrogen
by crops often is not higher than 30-40% (55).
Feedlot runoff and lagoon effluent can contribute to the nitrogen con-
centration in surface waters (52, 57). Land application of these liquids
would be a more appropriate disposal method than would discharge to
surface waters. Effluent from an anaerobic lagoon treating livestock
wastes was applied to the soil in summer and the following removals
were obtained - COD - 95%, phosphorus - 99%, and nitrogen - 80% (58).
No alteration in efficiency occurred with an effluent application rate
of 13.9 to 30.5 inches in three months. Good practices in disposing
of wastes on land such as plowing the waste material under as soon as
possible, use of soil injection systems, and avoidance of spreading on
frozen ground and snow will be beneficial in reducing the water pollution
problem especially when such practices are used in conjunction with
crop production.
Although there are several physical and chemical processes now available
for the removal of nitrogen, their applicability in treating animal
wastes has not been evaluated as yet. One of the methods, ammonia
stripping, has been examined in this study and is discussed in a sepa-
rate section.
One of the biological methods available for removal of nitrogen involves
the incorporation of the nitrogen of wastewaters into growing algal
cells in stabilization ponds. This method has been tried in the labora-
tory as well as in the field (59, 60). Chicken manure has been found
to have a high algal growth potential. This method may offer some
promise in regions having abundant sunshine provided a large quantity
of water is available for flushing and dilution purposes.
Practical Applications of Denitrification - Wastewaters containing
oxidized nitrogen and very little ammonia can be denitrified directly
if there is an adequate supply of hydrogen donors available. An indus-
trial waste containing nitrates and various organic substrates was
treated in a pilot plant study in which the denitrification was accom-
plished in a separate unit, ahead of a conventional activated sludge
basin (61). In the denitrification step a concomitant decrease in BOD
was reported with the disappearance of nitrates. The hydrogen donors
used in this process were a mixture of raw wastewater and activated
sludge. An average nitrate removal of 95% was accomplished.
In studying the feasibility of denitrifying an agricultural waste-
water containing relatively little organic matter, methanol was
used as a hydrogen donor (63). Three possible processes were indi-
cated: a) an anaerobic pond with detention times of several days
with recycling of the accumulated seed, b) an anaerobic activated
sludge system, and c) an anaerobic filter. It was reported
142
-------
that depending on the temperature, efficient nitrogen removals from
agricultural sub-surface drainage water can be accomplished with an
anaerobic filter having a hydraulic retention time of 0.5-2 hours (64).
An activated carbon pilot plant (0.3 MGD) treating secondary effluents
was operated with a two stage carbon adsorption sequence. The plant
achieved 80 and 92% removal of nitrogen in the respective stages. Most
of the denitrification occurred at a 10 minute detention time (66).
The feasibility of column denitrification has been verified in other
studies (67).
The practice of denitrification following nitrification has been tried
in laboratories, in pilot plants, and in large scale plants to remove
the oxidized forms of nitrogen from the sewage effluents. Two distinctly
different sets of conditions are necessary to control the processes of
nitrification and denitrification. Several schemes have been suggested
incorporating the separation of these processes.
Experience with activated sludge has shown that the endogenous reserve
materials in the bacteria can be available in adequate supply to serve
as hydrogen donors for the denitrification of a nitrified sewage. In
a laboratory study nitrification was accomplished in an aeration period
of two hours with a mixed liquor containing 5000-6000 mg/1 of suspended
solids. About 60-80% of the 25 mg/1 nitrogen in the settled sewage was
eliminated in a 2-5 hour contact period under anaerobic conditions (68).
The results of recent studies confirm that the endogenous reserves of
activated sludge will suffice to accomplish high nitrogen removals and
that there may be no need to supplement nitrified waste with exogenous
hydrogen donors (69, 70).
Other denitrification schemes have used raw sewage as a hydrogen donor.
In one such scheme, two reactor systems, one for nitrification and the
other for denitrification, were employed. The performance of such a
process was evaluated on laboratory and pilot scale models. It was
found that a raw waste:mixed liquor ratio of 1:5 removed practically
all the nitrate in 3 hours and that increasing the proportion of raw
waste did not appear to increase the rate of denitrification. Although
the addition of raw waste accomplished rapid rates of nitrate removal,
it lowered the overall removal of nitrogen because of the contribution
of unoxidized nitrogen by the addition of raw waste. This method could
only achieve a maximum overall nitrogen removal efficiency of about 70
percent (71).
In another study performed in Germany, the successful elimination of
nitrogen from sewage and the much stronger liquor from digesters was
achieved by first nitrifying the wastes and then denitrifying in a
separate vessel. A portion of the influent was diverted to the denitri-
fication reactor and was used as a hydrogen donor. It was reported that
60 percent of nitrogen was removed. No extra retention time was neces-
sary over that of a conventional activated sludge plant. The total
143
-------
decreased because of the addition of raw waste as a hydrogen donor
(72-74).
To facilitate controlled operation and rapid rates of nitrogen removal,
a three sludge system has been developed. Each sludge system has its
own recycling (75). The first system is a high rate activated sludge
and handles the bulk of the carbonaceous matter. The excess sludge
is wasted. The second sludge system receives and nitrifies the pre-
dominantly ammonia nitrogen feed from the first system. The nitrified
effluent from the second unit is sent to the third system, a stirred
anaerobic reactor. Methyl alcohol is added to the third unit as a
hydrogen donor in proportion to the concentration of nitrite and nitrate
nitrogen. Similar schemes with modifications have been in operation
(76, 77).
Another approach for the removal of nitrogen from existing plants has
been to combine nitrification and denitrification in the activated
sludge unit. This is accomplished by recycling controlled amounts of
mixed liquor from the effluent to the influent zone of aeration. Some
anaerobic conditions are created deliberately by reducing the aeration
in the influent zone. In this way, the nitrite and nitrate from the
recycled mixed liquor are denitrified with a concomitant oxidation of
the incoming load. The nitrogen gas formed is released during aeration
(78).
A flow sheet for the treatment of bovine wastes involving phosphorus
removal by addition of chemicals, and carbon and nitrogen removal by
oxidation and denitrification has been proposed. The authors found
that when properly employed, wet waste handling systems are more
effective to own and operate than systems designed to handle dry wastes
(79).
Microbiology and Biochemistry of Nitrification
Formation of Ammonia - About 5% of the total dry weight of poultry
manure is total nitrogen, which is primarily proteinaceous material
and uric acid. The degradation of these compounds occurs by hydrolysis
with the formation of lower molecular weight compounds. There are
many transformations that these nitrogenous materials can undergo in
biological systems. Only some of the important reactions that yield
ammonia will be discussed because of its importance as a substrate for
the nitrifying microorganisms.
Ammonia is derived from proteins through the formation of ami no acids
and their subsequent deamination. Deamination can occur by a) an
oxidative or reductive process or a combination of both, b) desatura-
tion, or c) hydrolysis with no net reduction or oxidation. Typical
examples of the above reactions follow.
144
-------
Oxidative deami nation - formation of a keto acid and ammonia
RCHNH2COOH + 1/2 02 + RCOCOOH + NH3 (53)
Desaturative deami nation - formation of an unsaturated acid
and ammonia
HOOCCH2CHNH2COOH + HOOCCH = CHCOOH + NH3 (54)
(aspartic acid) (fumaric acid)
Hydrolytic deami nation - formation of a-hydroxy acid and
ammonia
HOOCCH2CHNH2COOH + H20 -*• HOOCCH2CHOHCOOH + NH3 (55)
(aspartic acid) (malic acid)
Reductive deami nation - formation of a fatty acid or
dicarboxylic acid and ammonia
HOOCCH2CHNH2COOH + 2H+ + HOOCCH2CH2COOH + NH3 (56)
(aspartic acid) (succinic acid)
Mutual oxidation and reduction - Ammonia is formed in an inter-
molecular oxidation reaction also known as the Stickland reaction,
in which one ami no acid is oxi datively deami nated at the expense
of another which is reduced.
CH3CHNH2COOH + ZCH^COOH -»• 3CH3COOH + 3NH3 + C02 + 2H20 (57)
(alanine) (glycine) (acetic acid)
Uric acid decomposition - Uric acid is degraded either directly or
via allontoin to urea which in turn is degraded to ammonia by the
action of the enzyme urease.
All of these deami nation reactions and others are influenced by environ-
mental conditions. The oxidation reactions will occur predominantly
under aerobic conditions whereas the reductive reactions generally take
place under anaerobic conditions. The reductive deami nation may result
in the production of low molecular weight fatty acids.
In microbial systems where the C:N ratio is low, nitrogen can be miner-
alized following the production of ammonia. In activated sludge systems,
a number of nitrogenous compounds are known to be degraded with the
145
-------
formation of ammonia which is further oxidized to nitrate (80). Ammonia
is released in the biological waste treatment processes such as extended
aeration and aerobic digestion due to endogenous respiration which
results in lysis of the cells.
Nitrification - The conservation of nitrogen is an important consideration
in the field of agronomy and soil scientists have been studying various
facets of this subject for over a century. Our present day knowledge
on the transformation of nitrogen and the physiology of microorganisms
involved in these transformations stems primarily from their research.
Nitrification can be defined basically as the biological conversion of
nitrogen in inorganic or organic compounds from a reduced to a more
oxidized state. Often times in the field of water pollution control,
nitrification is referred to as a biological process in which ammonium
ions are oxidized initially to nitrite and then the nitrite is oxidized
further to nitrate.
Pasteur in 1862 suggested that the oxidation of ammonia might be micro-
biological. This suggestion was verified in classical studies with
sewage and soil (81). These studies showed that oxygen was essential
and that alkaline conditions favored nitrification. The ubiquity of
biological nitrification in soils has been well demonstrated (82).
The autotrophic nature of the bacteria responsible for nitrification in
earlier studies and the unavailability of special culture media to
isolate and study them made it very difficult for the pioneer micro-
biologists to obtain pure cultures of the organisms. The first suc-
cessful pure culture attempt was made by Winogradsky (83) who showed
that they would grow strictly on inorganic media.
For a long time it was considered that only autotrophic bacteria were
responsible for nitrification. It is now known that heterotrophic
bacteria, actinomycetes, and fungi also can bring about oxidation of
nitrogen to nitrite and nitrate.
The Nitrifying Organisms - Subsequent to the reports of Winogradsky (83)
and other enrichment and pure culture studies, several genera of nitri-
fying organisms were reported. Bergey's manual lists seven genera (84),
Nitrosomonas, Nitrosospira. Nitrosococcus. Nitrosocystis, Nitrosogloea,
Nitrobacter, and Nitrocystis.
Of these seven genera, only Nitrosomonas and Nitrobacter are generally
encountered in aquatic and soil ecosystems and are undoubtedly the
nitrifying autotrophs of importance. The other genera are rarely
reported and the validity of some of these is debatable because the
original strains may have been mixed with other organisms. Two more
new genera of obligate autotrophic nitrite oxidizing bacteria were
reported recently (85).
Although the nitrification process in nature is predominantly autotro-
phic, there are several heterotrophic organisms that can bring about
the oxidation of nitrogen. A detailed list of these organisms is
146
-------
available (86). Outstanding in this regard is Aspergillus flavus which
forms substantial amounts of nitrate nitrogen from ami no compounds
(87, 88).
Physiology of the Nitrifying Organisms - The important nitrifying
organisms are obligate autotrophs and use the energy derived from the
oxidation of ammonium and nitrite for synthesis. It has been reported
that organic compounds have an inhibitory effect on the growth of
nitrifying organisms (89) but the claim that the organic compounds in
general are inhibitors is perhaps over emphasized. It was reported
that a species of Nitrosomonas isolated from farm yard manure oxidized
ammonia and exhibited "exceptional resistance" to 0.5 M glucose, formate,
acetate, glycerol and succinate (90). The behavior of the nitrifying
population probably differs significantly in pure culture and in the
presence of organic matter of an ecosystem. It was shown that peptone
was far less inhibitory in sand than in broth media (91). In soil
itself nitrification was inhibited much less in the presence of organic
matter than in pure culture media (92, 93). The occurrence of nitri-
fication in soils containing organic matter, in trickling filters and
activated sludge tanks of sewage treatment plants, and in compost piles
(94, 99) bears testimony to the fact that the process takes place freely
in natural ecosystems containing varied degrees of organic matter.
Considerable attention has been given to the nutrition of the nitri-
fying organisms. The majority of the nutritional studies have been
conducted using Nitrosomonas europea. The nitrogen of amino acids,
amides, proteins, or urea is not oxidized by Nitrosomonas europea,
although the ammonia formed by prior deamination in certain purines is
utilized and converted to nitrite (100). Several other carbon sources
have been investigated (101) but there is no evidence that they are
used as sources for either carbon or energy for growth. Nitrobacter
agilis is reported not to be a strict autotroph (102). The organisms
require magnesium, phosphorus (103). Copper and iron were found to
have a good response in the growth of these organisms (104, 105).
Several culture media were formulated considering the nutritional
aspects of the bacteria (106 - 108) and the organisms can be enumerated
using a MPN technique (109) as well as by a membrane filter technique (110).
The generation time for the nitrifying organisms is longer than that of
heterotrophs. An apparent generation time for ammonium oxidizers in soil
was calculated to be thirty hours (111).
Inhibitors of Nitrification - From an agronomic point of view, conserva-
tion of fertilizer nitrogen is important. To inhibit the production of
nitrites and nitrates from ammonia and hence inhibit the loss of nitrogen
by denitrification, a number of potential chemical inhibitors have been
evaluated (112).
In the determination of BOD of wastewater, nitrification can induce an
error unless a correction is made. To inhibit the nitrification during
the BOD test, several inhibitors were evaluated, and modified procedures
for measuring the BOD have been recommended (113-116).
147
-------
Biochemistry of Nitrification - There have been several reviews on the
metabolism of the autotrophic nitrifiers (112, 117, 118). The current
status on the biochemistry of heterotrophic nitrification was reviewed
in a recent dissertation (86). The general metabolic schemes postulated
for the two types of nitrogen metabolism are given below.
a) Autotrophic Nitrification - This pathway represents the change of
nitrogen valence from -3 to +5, specifically the conversion of ammonia
to nitrate involving! the release of 8 electrons. The postulated path-
way (Equation 58) was reported four decades ago and at the present time
there is neither adequate confirmation nor a satisfactory alternative
pathway.
NH4 •* NH2OH •> HNO -»•
hydroxyl (nitroxyl)
ami ne
NH(OH)2 + HN02 -> ONH(OH)2 + HN03 (58)
dihydroxy (nitrous acid)
ammonia
Values of the free energy change in the oxidation of ammonium to nitrite
and nitrite to nitrate were reported to be in the range of -65.5 to
-84 KCal per mole of ammonium and -17.5 to -20 KCal per mole of nitrite
oxidized respectively (119). Thus the nitrosomonads obtain more energy
than the nitrobacters. Assuming the efficiency of cell synthesis is
the same, more nitrosomonads are formed than nitrobacters per unit of
ammonium undergoing nitrification in an ecosystem. In other words, the
nitrobacters should approximately utilize three times more substrate
than the nitrosomonads for synthesizing the same amount of cell mass.
This is why insignificant accumulations of nitrite occur in an ecosystem
nitrifying under no adverse environmental conditions.
b) Heterotrophic Nitrification - The pathway for the oxidation of nitro-
gen combined in organic compounds such as amines or amides can be postu-
lated as follows:
RNH2 + RNHOH + RNO + RN02 -* N02 + NO., (59)
R = NOH nitroso nitro (nitrite) (nitrate)
substituted compound compound
hydroxylamine
An experimental study dealing with the elucidation of a pathway for
heterotrophic nitrification using natural ecosystems and a pure culture
has been reported (86). In this study heterotrophic nitrification was
shown to occur in sewage, lake and river waters.
148
-------
Factors Affecting Nitrification - Although the soil scientists have
been studying the process of nitrification and the factors that affect
it for the past century, the interest on the subject in the wastewater
field is of more recent origin. During the 1930's, experiences indi-
cated that well nitrified effluents from waste treatment plants resisted
putrefaction and the emphasis was to achieve a highly nitrified effluent.
However, with the advent of the biochemical oxygen demand test, sanitary
engineers have tended to design treatment plants for efficient removal
of BOD with comparatively less expense by deliberately avoiding the
nitrification of the effluent. Another reason for lack of interest in
producing a nitrified effluent has been sludge rising in secondary
clarifiers due to denitrification. Elimination of nitrification was
considered a cure for this problem.
The implication of the nitrogenous oxygen demand was well understood
with respect to its effect on the receiving stream. Nevertheless, the
relatively un-nitrified secondary effluents from waste treatment plants
are discharged into streams ignoring these effects, namely a) that the
nitrogenous oxygen demand will be exerted in streams thereby decreasing
the dissolved oxygen of streams and b) the resultant products of
nitrification may trigger algal blooms. The latter also could be a
problem when well nitrified effluents are discharged into streams.
There is now an awareness to remove these nutrients before discharge.
This approach also reduces the cost of disinfection of the water and
treated effluents.
Irrespective of whether the objective is to inhibit or promote nitri-
fication, a basic knowledge of the factors that govern this process is
of value. A consideration of these factors is given below.
a) Microorganisms - Since the process of nitrification depends upon the
metabolism of a certain group of highly specialized aerobic organisms,
it is imperative that these organisms should be present in adequate
numbers. High rates of aeration alone have not achieved nitrification
with liquid poultry manure due to the absence of nitrifying organisms
(120). However, by seeding the poultry manure with soil, nitrification
was achieved.
Similar results were obtained by inoculating sewage with 1% by volume of
activated sludge; nitrification proceeded from the first day. In a
parallel run unseeded sewage did not show signs of nitrification till
the fifth day (121). This suggests that nitrification can be made to
proceed without any lag provided the system is seeded adequately with
nitrifiers initially. It was also reported that the rate of nitrifica-
tion increased with increased concentration of suspended solids in an
activated sludge unit (122). The distinct separation between carbona-
ceous and nitrogenous oxygen demand stage of a BOD curve is due to the
fact that nitrifying bacteria are not usually present in adequate numbers
initially and that their effect is seen after a lag period during which
they are actively multiplying (123). The protozoa! population associated
149
-------
with activated sludge has been considered as an important factor in
nitrification (122).
Although there is some relevant information on the implications of the
nitrification process in waste treatment, information on its kinetics
has been relatively sparse till recently. To accomplish consistent
nitrification in a waste treatment system it is necessary to maintain
a nitrifying population in the system. This can be accomplished in a
continuous flow system only if the rate of growth of the nitrifying
population is greater than the rate at which they are removed as the
excess sludge. It is necessary to provide a certain minimum detention
time for the multiplication of the organisms. At very low detention
times the organisms in the aeration basin are diluted and washed from
the system before they can multiply. Based on the above conditions,
adequate nitrification has been reported with a solids retention time
(SRT) of 2-4 days (96). In order to attain a higher degree of nitri-
fication consistently, a SRT of 4 days has been found necessary (95, 98).
b) Dissolved Oxygen - Since the nitrifying organisms are aerobic, a
waste stabilization system having an adequate dissolved oxygen (DO) can
support nitirification provided other conditions are non-inhibitory.
It is generally accepted that an increase in the DO results in the
increase of nitrification up to a certain level beyond which oxygen
concentration has little effect on nitrification. An experiment to
determine the important factors affecting the formation of nitrate
implied that the dissolved oxygen has the most significant effect
followed by sewage ammonia content, BOD, MLVS, and aeration time on
the formation of nitrate (125).
A pilot plant study indicated that a DO concentration greater than 0.5
mg/1 had no apparent inhibitory effect on nitrification (126) while
other studies reported that nitrification could be accomplished at DO
concentrations less than 0.5 mg/1 (127, 128). Relatively poor nitri-
fication has resulted at DO levels less than 1.0 mg/1 in comparison to
the nitrification acheived in plants operated at a mixed liquor DO of
4 and 7 mg/1 (129). At least 1 mg/1 of DO in aeration basins of con-
ventional plants and perhaps slightly higher concentrations in high
rate plants should be maintained (96). The optimum nitrite concentra-
tion for nitrite oxidation by Nitrobacter was reduced as the oxygen
tension was lowered (130). Oxygen requirements for the oxidation of
ammonia and nitrite have been found to be 3.22 mg/1 and 1.11 mg/1 of
oxygen utilized to oxidize 1 mg/1 of NH4~N and NOp-N respectively (131).
Results from pure culture studies have been different from those observed
in natural ecosystems. Activated sludge containing detectable ammonia
when left aerobically or anaerobically for four hours lost little or no
nitrifying ability whereas Nitrosomonas cultures have quickly lost their
activity under similar conditions (132). Nitrobacter also lost its
activity when it was starved of nitrite (133). In samples of Thames
River water in which ammonia was oxidized completely, it was found that
150
-------
Nitrosomonas lost its activity and after 4-5 days without ammonia only
34-57% of the population survived (132).
c) Temperature - From pure culture studies, the optimum growth of nitri-
fiers was found to occur between 30 and 36°C (112). Nitrobacter was
found to grow optimally at 42°C (136). An exposure of 10 minutes at
53-55°C and at 56-58°C killed Nitrosomonas and Nitrobacter respectively
(137). These bacteria are known to be very resistent to drying con-
ditions in soils but not to drying in liquid cultures. They can be
maintained for two years if water is added from time to time (103).
Alternate drying and wetting of soil stimulates microbial activity
resulting in greater mineralization of nitrogen (138). A comparison
of nitrification using heated and unheated soils indicated that heated
soils increased the mineralization of nitrogen during alternate drying
and wetting cycles (139).
Laboratory activated sludge studies indicated that the rate of nitri-
fication increased throughout the range of 5-35°C and that the rate of
ammonia oxidized was much higher at 31.5°C than at 8°C (126). While
results from one pilot plant study indicated that it was possible to
maintain nitrification at 8°C (76), another study revealed that nitrifi-
cation did not develop below 10 C (140). The growth rate constant for
Nitrosomonas in activated sludge roughly doubled for each 10°C increase
in the range of 6-25°C (96). Thus a 10°C drop in temeprature would
roughly double the aeration period for the same degree of nitrification
if the nitrifying organism population in the mixed liquor was not
increased proportionately. This temperature effect is the reason why
treatment plants do not nitrify well in winter time. It was observed
that very little nitrification was achieved at temperatures below about
6°C even when surplus sludge was not removed (96). However, the adapta-
bility of nitrifiers to low temperatures in soils has been reported
(141, 142). Nitrification was not found to be significant at 50°C in
laboratory units (143).
d) pH - The optimum pH for the growth of the nitrifiers is not sharply
defined but in pure cultures it has been shown to be generally on the
alkaline side of neutrality. A plot of the oxygen uptake against
various pH values resulted in plateaus occurring at pH 8.5-8.8 for
Nitrosomonas and at 8.3-9.3 for Nitrobacter. On either side of the
plateau the curves dropped off steeply (89). Different pH optima have
been reported in other studies (144). The influence of pH on oxidation
rate of a cell free extract of Nitrobacter was found to be different
than the oxidation rate exerted by intact cells (146). From these
studies the pH value for the optimum oxidation rate was calculated as
7.7 for cell free extract and 8.2 for intact cells at 32°C.
The effect of pH on the nitrification of sewage and activated sludge was
reported to be variable and the optimum range appears to be between
7.5 and 8.5. The optimum pH for nitrification in an activated sludge
study was found to be 8.4. Ninety percent of the maximum rate occurred
151
-------
in the range of 7.8-8.9 and outside the ranges of 7.0 to 9.8 less than
50% of the optimum rate occurred (126).
During nitrification, hydrogen ions are produced and as a result the pH
drops. When more than 60 mg/1 of nitrogen was oxidized with an acti-
vated sludge treating domestic sewage, pH values as low as 5-5.5 resulted.
The high acid concentration caused bulking problems and reduced the
efficiency of BOD removal (132). Low pH values as a result of nitrifi-
cation were also noted in compost piles and in soils (97, 112).
In a pilot plant study, the feasibility of developing a nitrifying flora
in a separate aeration step was studied. It was not possible to maintain
an active nitrifying flora below a pH range of 8.3-8.5 (140). However,
in another study made in Europe, it was found that the optimum pH for
nitrification was approximately 7 and at pH values less than 6.9, it
was inhibited (74).
e) Ammonia and Nitrite Concentration - Although NH4+ is the energy source
for the nitrifiers, excessive amounts can inhibit the growth of these
bacteria. Ammonia is more toxic to Nitrobacter than to Nitrosomonas
and it is reported that the toxicity to the former species is more due
to undissociated (free) ammonia than ammonium concentration (147). A
concentration of 0.0005% of NhL retarded nitrite oxidation considerably
and a level of 0.015% NH3 was sufficient to stop the nitrification (105,
148). An inhibition of 70% of the nitrification at a concentration of
0.001M NH3 at pH 9.5 was reported (89).
The application of large amounts of ammonia forming fertilizers to soil
can lead to the accumulation of nitrite nitrogen in the soil. In soil
perfusion experiments the nitrite nitrogen accumulation was proportional
to the initial ammonium nitrogen applied (111). Irrespective its con-
centration, the N02-N formed was oxidized after all the NH.-N was oxi-
dized. It was concluded from the study that high NH^-N concentrations
can cause the N02-N accumulation in soils of high pH because of the spe-
cific toxicity of ammonia for Nitrobacter.
Nitrite reduces the activity of microorganisms at low pH levels (150).
Nitrobacter did not grow in a medium containing 1100 ppm NO?-N; however,
it possessed the mechanism to adapt to higher concentrations of nitrite
and withstand its toxic action when the nitrite was added in steps (147).
Even a very low concentration of undissociated nitrous acid was lethal
to Nitrobacter (146).
DENITRIFICATION
Denitrification refers to the reduction of nitrite and nitrate culmi-
nating in the liberation of gaseous end products such as molecular
152
-------
nitrogen and/or nitrous oxide (N^O). Occasionally nitric oxide (NO) also
is found as an end product. Denftrifi cation can take place chemically
without the aid of microorganisms, as well as biologically.
Chemical Denitrifi cation - There are a few known mechanisms by which
nitrogen can be lost chemically. Under suitable conditions nitrous
acid reacts with amino acids or primary amines to yield molecular
nitrogen according to the Van Slyke reaction.
RNH2 + HN02 •> RON + H20 + N2* (60)
This reaction occurs predominantly at pH values less than 5. Conditions
for the formation of nitrite or nitrate are not favorable at these pH
values (151). Consequently losses of nitrogen by this mechanism may not
be significant in natural ecosystems.
In a reaction similar to the one above, ammonia released from urea or
uric acid may react with nitrous acid to yield molecular nitrogen:
NH3 + HN02 -> N2t + 2H20 (61)
Ammonium nitrite was suggested to be formed instantaneously by the reaction
between ammonia and nitrous acid and then degraded to nitrogen gas. Much
of the loss of nitrogen in aerobic systems (aerobic denitrifi cation) was
claimed to be the result of the formation and decomposition of ammonium
nitrite (152). However, several other researchers revealed evidence
against such a mechanism (153). It appears that this reaction (Equa-
tion 61) is uncommon in natural ecosystems.
Under acid conditions nitrous acid can undergo the following changes:
3HN02 + 2NO + HN03 + H20 (62)
nitric
oxide
Nitric oxide can be oxidized chemically to NO,,.
2NO + 02 -»• 2N02 (63)
nitrogen
dioxide
153
-------
The nitrogen dioxide may react with water to form nitric acid.
3N02 + H20 + 2HN03 + NO (64)
2N02 + H20+HN03 + HN02 (65)
It is likely that the nitric oxide formed in Equation 64 and the nitrous
acid formed in Equation 65 may react as in Equations 62 or 63. If this
sequence of events takes place, no significant losses of nitrogen will
result, but increases of nitrate will occur. The contribution of nitro-
gen losses by the above pathway in waste systems are minimal because
these reactions take place under acid conditions, about pH 4. It is
unlikely that such low pH values will be encountered in biological
waste treatment systems.
Microbial Denitrification - Microbial denitrification takes place under
anaerobic conditions where nitrites and nitrates are used as terminal
hydrogen acceptors in place of molecular oxygen. Fundamental knowledge
on this subject primarily stems from the observations reported on the
denitrification in soils. Earlier studies indicated that denitrifica-
tion takes place rapidly in highly organic environments such as manured
soils and that it progresses slowly in well aerated soils (118).
Denitrification is brought about by facultative bacteria. The genera
of Pseudomonas and Serratia are the dominant ones that bring about
denitrification in soil. Most of the active members belong to the
genera of Pseudomonas, Achromobacter, Bacillus, and Micrococcus.
Species of the genera Chromobacterium, Mycoplana, Serratia, and Vibrio
are known to catalyze the reduction of nitrate (112). Some chemo-
autotrophs are known to reduce nitrate. For example, Thiobacillus
denitrificans uses elemental sulfur or thiosulfate as its energy source,
and nitrate is converted to gaseous nitrogen. Micrococcus denitrificans
grows facultatively with either organic compounds or molecular hydrogen
as an energy source at the expense of oxygen or nitrate as an electron
acceptor.
There are three major types of nitrate reduction known to occur in the
microbial world (154).
a) Nitrate Assimilation - The reduction of nitrate is required for the
building of cell protein or as a substitute for the reduction of oxygen
in conventional aerobic metabolism. The primary products are not gaseous
in nature. Ammonium is the end product and enters the pathways leading
to the synthesis of protein. The whole process of nitrate reduction
utilized for the synthesis of protein is known as assimilatory nitrate
reduction.
154
-------
b) Incidental Dissimilatory Nitrate Reduction - This kind of a reaction
is found in some kinds of bacteria normally aerobic in nature, but which
can utilize nitrate in place of oxygen as a hydrogen acceptor. The end
products are nitrite or ammonium nitrogen.
c) True Dissimilatory Reduction - This type of reaction is also of impor-
tance to the cell since both nitrite and nitrate are utilized as electron
acceptors without the accumulation of toxic concentrations of end products
such as nitrite and ammonia. The actual end products are gaseous
nitrogen, nitrous oxide, or nitric oxide. In nature, formation of NH.
as an ultimate product of reduction by true dissimilatory reduction is
known to occur (154).
The pathway for the nitrate reduction and denitrification is (112):
2HN03 + 2HN02 + [HON = NOH] + 2NH2OH (66)
hyponitrite hydroxyl-
ami ne
* 4- +
N 0 -*- N ?NH
iio" '"1 L.nn«5
From this pathway the following equations can be written
2N03" + 10H+ N2 + 4H20 + 20H" (67)
2N02" + 6H+ + N2 + 2H20 + 20H" (68)
N20 + 2H+ -> N2 + H20 (69)
In truly dissimilatory nitrate and nitrite reduction, the nitrate and
nitrite are used as hydrogen acceptors and not as a source of oxygen for
the microbial processes. A conceptual misunderstanding with respect to
this role of nitrates and nitrites appears to have existed in the field
of sanitary engineering (70-71).
Molecular nitrogen is generally the major end product of denitrifi cation.
The relative proportions of N2 and N20 are reported to depend on the pH
of the denitrifying system. Nitrous oxide was readily reduced to N2
above pH 7, but its reduction was strongly inhibited below pH 6 (155).
An increase in the amount of N^O was observed at higher ammonium
155
-------
concentration in soil. When all the nitrogen was present as nitrate,
the losses were entirely in the form of gaseous nitrogen (156).
Denitrification in Haste Treatment Systems - Losses of nitrogen from
waste treatment systems have been observed for a long time. As early
as 1886, 85 to 95% nitrogen losses from sewage disposal plants were
reported (157). Denitrification in stabilization ponds has occurred.
The seepage of nitrate containing pond water through anaerobic pond
sediments results in denitrification (158). In oxidation ditches used
for the treatment of poultry waste, denitrification has been reported
even though bacterial nitrification was known to occur simultaneously.
A total nitrogen loss of about 30% was presumed due to denitrification
in the settled particulates and the inner portion of the microbial floe
(99). Ammonia in oxidation ditch mixed liquor was converted to nitrate
but no nitrate was measured in the effluent indicating nitrogen loss
due to denitrification (159).
The losses of nitrogen due to denitrification also were reported to occur
in actively nitrifying activated sludge and trickling filter plants (98,
158, 160-162). The losses of nitrogen occurring at low organic loading
rates were in the range of 50-55% of the total nitrogen. Microbial
denitrification was an essential factor for these losses.
The important factors that govern the denitrification in an ecosystem
are a) organic matter, b) oxygen tension, c) pH and d) temperature.
a) Organic Matter - Substrates that can act as hydrogen donors are
necessary for the denitrification of oxidized nitrogen to occur. These
are primarily oxidizable organic compounds and act as energy sources for
the denitrifying population. The degree and rate of denitrification
depends on the ease with which these hydrogen donors are degraded by
microorganisms (163). In soils, bacteria of the rhizosphere are reported
to reduce the oxygen tension and cause denitrification. The consumption
of the root exudates as hydrogen donors was implicated (164). Under
such situations, denitrification in soils may take place even if no
intentional addition of organic matter occurs. However, the addition
of organic matter to a soil already treated with water increased the
nitrogen losses significantly (165).
An agricultural wastewater containing high nitrates and relatively low
amounts of organic matter was successfully denitrified by the addition
of methanol as a hydrogen donor (63). Several organic compounds were
evaluated in this study for their ability to serve as hydrogen donors.
Methanol was the most economical and easy to use to achieve denitrifi-
cation. An equation for the quantity of methanol required to reduce
the nitrate, nitrite and dissolved oxygen was developed:
Cm = 2.47 N03N + 1.53 N02N +0.87 DO (70)
156
-------
Cm is the methanol required. The use of methanol as a hydrogen donor
for the removal of nitrogen from municipal waste also was studied by
several other investigators using a number of anaerobic processes (67,
166). While these researchers advocated the use of methanol as an
external electron donor, it was reported that by manipulating the
activated sludge process itself, it was possible to achieve denitrifica-
tion by using the endogenous reserve material of bacteria contained in
the activated sludge without the addition of any exogenous hydrogen
donors (98). A similar approach was reported to be effective for
removing nitrate from sewage (167). Although good denitrification was
achieved using the endogenous reserves with an activated sludge con-
taining 4000 mg/1 of MLSS at 24°C, the formation of N~ did not occur at
11-14°C at the same concentration of MLSS unless an additional hydrogen
donor was added (168). It was also found that addition of glucose and
nutrient broth increased the rates of denitrification.
b) Oxygen Tension - The presence of dissolved oxygen is detrimental to
the process of denitrification. Nitrate was not reduced at dissolved
oxygen concentrations of about 0.2-0.4 mg/1 (169, 170). On the other
hand, nitrite was reported to be reduced in the presence of oxygen at
a concentration as high as 8 mg/1 (171, 172). This appears to be
possible from a thermodynamic viewpoint (173), although this possibility
has not been examined in detail.
In pure culture studies, there have been conflicting reports in which
denitrification was observed in environments with adequate aeration
(152, 174-176), and frequently the term "aerobic denitrification" is
used to explain such results. The explanation may be that the micro-
environment of the organism was depleted of dissolved oxygen due to the
rapid metabolism of some of the organic matter. Under these conditions,
the cells may have utilized nitrate or nitrite present in its immediate
vicinity as a hydrogen acceptor, although there may have been a con-
siderable amount of residual dissolved oxygen in the culture medium.
The losses in nitrogen under aerobic conditions reported in waste treat-
ment plants were presumably due to such conditions.
In a study using soil columns, the loss in nitrogen concentration was
correlated with the decrease in redox potential, oxygen content of the
soil solution, and oxygen levels in the soil atmosphere and with increases
in soluble iron and manganese. It was concluded that it was feasible
to remove nitrate nitrogen from agricultural effluents by subjecting
them to denitrification (165).
Oxygen acted as a powerful inhibitor at pH values of 6.5-7.0 for both
nitrate and nitrite reduction (177). The inhibition was less at lower
pH values. Efficient denitrification was only possible under strict
anaerobic conditions. A ratio of 0.91-1.28:1 of oxygen removed to BOD
removed was reported as an estimate of the quantity of oxygen demanding
substance needed to remove dissolved oxygen (71). Equation 70 predicts
157
-------
the methanol requirement for the reduction of dissolved oxygen present
in a denitrification reactor.
c) PH - It is generally considered that active denitrification takes
place under neutral or slightly alkaline conditions and that it is not
favored under acid conditions (178-179). Contrary to these beliefs, it
has been reported that there was no correlation between pH and other
denitrification parameters (180). The inconsistencies in the literature
are likely due to the differences in the behavior of the various microbes
to the complex microbial interactions in the mixed culture studies.
The toxicity due to nitrite appeared to inhibit denitrification in acid
conditions (181). The relative proportion of gaseous end products and
their dependence on pH was noted earlier (155).
d) Temperature - Like any biochemical process, denitrification also
shows a temperature dependency. The optimum temperature reported for
the denitrification process was high, in the range of 60-65°C (163, 179).
The proportion of N20 was higher at lower temperatures with N2 being
the principle end product at higher temperatures (182). When the range
of 10-37°C was studied, maximum denitrification took place at 30°C with
the denitrifiers isolated from soil.
e) Initial Nitrate and Nitrite Concentrations - In soils, the initial
nitrate concentration was reported to have no influence on the denitri-
fication rate (155, 179, 181). However, with a municipal waste, the
denitrification rate was reported to increase with an increase in
nitrate concentration (95). A 0.1% nitrate concentration was optimal
for denitrification, but the same concentration of nitrite was inhibi-
tory to some soil bacterial isolates and a 0.2% nitrite concentration
suppressed growth almost entirely (182). If a large amount of unused
nitrate remained in the system, the resultant nitrite tended to accumulate
and could be toxic.
f) Redox Potential - An interdependence of redox potential and denitri-
fication was demonstrated by several investigators. In one study it was
reported that above 350 mv nitrates would accumulate and below 320 mv,
they would disappear (183). The nitrate became unstable at 338 mv (184)
and in a recent study it was found that rapid losses of nitrogen via
denitrification occurred when the redox potential was dropped to 300 mv
or below (165). Contrary to the above reports, redox potential was not
found to be a limiting factor in the reduction of nitrite (172).
OBJECTIVES AND METHODS
Objectives of the Study
The general objective of this phase of the project was to investigate
the feasibility of a microbial nitrification-denitrifi cation process
for the removal of nitrogen in animal wastes. Poultry manure wastes
were used to evaluate the feasibility of the process.
158
-------
Investigations on the removal of nitrogen from municipal wastewaters have
demonstrated the feasibility of nitrogen removal by nitrification of the
treated wastes followed by denitrification. The intent of this research
was to utilize the fundamentals of the microbial processes involved and
the available engineering knowledge of the use of the processes with other
wastes in developing suitable approaches that could be incorporated in
animal waste management systems.
>:
To evaluate the fundamental parameters that influence microbial nitri-
fication and denitrification of poultry wastes, extensive laboratory
studies were undertaken. The objectives of these studies were to study:
a) the need for seed organisms in the nitrification process
b) the effect of organic loading on nitrification
c) the effect of solids retention time (SRT) on the nitrification process
d) the effect of varying NH.-N, NCL-N, and NOo-N on the nitrification
process
e) the effect of pH on both the nitrification and denitrification
processes
f) the effect of dilute and concentrated mixed liquors on both the
nitrification and denitrification processes
g) the maximum degree of nitrification that will occur
h) temperature variations on the denitrification process
i) the type and magnitude of hydrogen donors required in denitrification
j) the quality of gases produced during denitrification
k) the possibility of a cyclic nitrification-denitrification process
Materials and Methods
Feed Suspension - Suspensions of feed were made with fresh poultry manure
obtained from poultry production units at Cornell University. The required
concentrations of the feed were made by suspending the requisite amount
of poultry manure in distilled water. The mixture was blended in a
Waring blender and filtered through a single layer of cheesecloth to
remove the feathers and the large particulate matter. The material
retained on the cheesecloth was washed with distilled water to recover
most of the soluble matter. The filtered suspensions were diluted to
the required volume with distilled water.
Mineral Salts Solution - A mineral salts solution was used in certain
experiments to resuspend centrifuged mixed liquor solids. The salt
159
-------
solution had the following concentration: 250 rng/1 of MgS04-7H20, 10 mg/1
of FeS04, 10 mg/1 of MnS04> 10 mg/1 of CaCl2-2H20, 100 ml tap water,
10 ml of 0.1M phosphate buffer solution, and 900 ml of distilled water.
The final pH of the solution was adjusted to 6.7.
Continuous Flow Units - The continuous flow nitrification.experiments
were conducted in continuous feed, constant overflow plexiglass units
using diffused aeration. Poultry waste feed suspensions of different
concentrations were used in the separate units and at different times
to vary the loading on the units. The aeration rates used in the
various units were adequate and high dissolved oxygen residuals were
maintained. The various forms of nitrogen, the nitrogen balance,
organic removal, and solids balance were monitored routinely- The
laboratory units were in a 20°C constant temperature walk-in chamber.
Poultry manure suspensions contained in 8 liter aspirator bottles were
held in a low temperature (4°C) water bath while they were fed to the
nitrification units (Figure 51) by means of a Dial-a-Pump. The desired
flow rates of the feeds were reasonably well maintained by keeping the
pump channels clean and by mechanically mixing the feed in the reservoirs.
During the initial phase of this study some of the units were operated
with internal solids recycling and without intentional solids wasting
For most of the experiments they were operated as completely mixed units
without solids recycling. In all the units air was supplied by an air
compressor and was humidified by passing through a water reservoir. A
brief description of the various units operated during the entire period
of this study follows.
a) Units Without Recycling of Solids
Unit A: One liter of a poultry manure suspension was placed in a 10 liter
plexiglass unit, and seeded with an aqueous extract of garden soil
obtained by filtering a soil suspension (lOOg in 250 ml water) through
a coarse sieve. The unit was fed 1 liter of poultry manure per day.
Unit A': The unit was started by combining 1 liter of an actively
nitirfying activated sludge and 9 liters of poultry wastes. One liter
of poultry manure suspension was fed every day.
Unit B: Unit B was operated exactly the same way as Unit A except that
it was not seeded with any material.
Unit B': The above Unit B was seeded with one liter of an actively
nitrifying activated sludge mixed liquor after 3 weeks of operation
without such seed.
Unit C: Mixed liquor, 0.5 liters, from the unit A1 was used as seed
for this unit. For 4 days one liter of poultry manure suspension was
160
-------
REFRIGERATED
WATER BATH
REMOVABLE
BAFFLE
EFFLUENT
ROTAMETER
FIGURE 51
EQUIPMENT USED FOR
CONTINUOUS FLOU NITRIFICATION STUDIES
161
-------
fed to this unit. On the 4th day, 8.5 liters of a poultry manure sus-
pension was added to the unit with the subsequent feed being one liter
of a poultry manure suspension per day.
Units D'. E1, F1. G1: All these units were operated in the same manner.
The units were filled with poultry manure suspensions of different con-
centrations and seeded with an actively nitrifying population at a ratio
of 10:1 (feed:seed) initially. The units were then fed with different
volumes of feed depending on the SRT in the units. Units E1, F1, and G1
were operated at various SRT values from about 1 to 20 days in this study.
b) Units with Recycling of Solids
Units D, E. and F: Plexiglass units equipped with baffles were used.
The start up, feeding of the units and other operational details were
similar to those for Units B, B1, and C. In these units, solids were
not wasted deliberately. Each unit had a separate mixing and solids
separation section. The overflow from the mixing section went to the
solids separation section where quiescent conditions and sedimentation
took place. The settled solids automatically were recycled to the
mixing section. Non-settleable solids were discharged in the effluent
from the sedimentation section. The solids in the mixed liquor of these
units settled readily and there were few solids in the final effluent.
Poultry manure suspensions of different concentrations were used as feed.
Batch Units - Nitrification - Suspensions of poultry manure at various
concentrations were made as described above. Different volumes of
poultry manure suspensions placed in 8 liter aspirator bottles were
seeded with a known volume of a highly nitrifying activated sludge
developed on poultry manure. The units were aerated with humidified air
and high levels of dissolved oxygen were maintained. Although humidi-
fied air was used for aeration, losses in the volume of mixed liquor
occurred due to evaporation and as a result the mixed liquor became
concentrated. Before samples were withdrawn for analysis, distilled
water was added to compensate for the losses due to evaporation.
The installation of a manometer on the side of the aspirator (Figure 52)
facilitated the addition of an exact amount of distilled water needed
for evaporation correction. The proper water level in each unit was
marked on the nitrification reactors and the manometers after opening
the pinch clamp. The pinch clamp was used to avoid quiescent conditions
and possible denitrification in the manometer section.
Before taking a sample, the pinch clamp was opened, distilled water added
to adjust the liquid level to the mark, any liquid in the manometer
returned to the reactor by tilting the reactor, the pinch clamp placed
in position, the liquid was thoroughly mixed, and the sample taken. This
procedure was followed prior to each sampling which was approximately
once each day.
Denitrification Reactors - All denitrification experiments were done in
batch units in which the change in nitrites, nitrates and other parameters
162
-------
CTl
OO
AIR-
1
P7^ F- —t *"^
& 'i.1'
sSWfJ
NITROGEN GAS
SERUM CAP
MANOMETER
DIFFUSER
PINCH CLAMP MAGNETIC STIRRER
NITRIFICATION DENITRIFICATION
BATCH REACTOR
FIGURE 52
SCHEMATIC OF UNITS USED FOR
BATCH NITRIFICATION AND DENITRIFICATION STUDIES
-------
were measured as time increased. Aspirator bottles (560 ml capacity),
containing magnetic mixing bars, and fitted with a pinch clamp were used.
Nitrified mixed liquor from various continuous flow nitrification reactors
were used in the denitrification experiments. The nitrified waste was
deoxygenated by passing N2 for 15 minutes and was transferred to these
reactors under an atmosphere of nitrogen. The reactors were closed
tightly with serum caps, covered with parafilm, and placed on magnetic
bases (Figure 52).
The reactor contents were gently stirred. Periodically samples were
withdrawn for analyses under an atmosphere of nitrogen in the following
manner. The pinch clamp was opened and a slight pressure was applied
simultaneously on the reactor contents by passing N2 through a hypodermic
needle inserted in the serum cap. Usually during the initial phases of
the denitrification study, this procedure was not found to be necessary
since there was adequate gas pressure in the reactor. As the pinch
clamp was opened, sufficient quantity of the sample was obtained for
analyses.
Methods - Total solids, volatile solids, and BOD were determined as
described in Standard Methods (12).
Suspended solids were determined by filtering a known volume of the
sample through glass fiber filter papers. The weight of the dry solids
retained on the filter paper was used as an estimate of the suspended
solids. A considerable length of time was generally taken for the
filtration of the realtively concentrated samples. In such situations,
a part of the weight of the suspended solids may have included some
dissolved solids.
The pH of the sample was measured with a Corning pH meter.
Ammonium nitrogen was determined by the distillation method (12). The
sample was buffered with phosphate buffer to pH 7.4 and was distilled in
a micro Kjeldahl distillation apparatus. The distillate was collected
into boric acid and titrated with KH(I03)2.
Total Kjeldahl nitrogen was determined by a micro Kjeldahl method (191).
Nitrite nitrogen was determined by a diazotization method using N-l-
napthyl ethylene diamine dihydrochloride (192).
Nitrate nitrogen was determined by the PDSA method as described in
Standard Methods but with some modification. The clarification of the
sample with A1(OH)3 was omitted. Clarification was achieved by a very
high sample dilution which was found to be necessary to obtain the range
of N03-N for suitable determinations. N02-N in the sample was eliminated
after the chlorides were removed. A neutral aliquot of the diluted
sample was adjusted to pH 2 by the addition of sulfamic acid crystals.
164
-------
The sample was stirred for five minutes. The sample pH was readjusted
to 7, and N03-N was determined. The volume of alkali added in adjusting
the pH was recorded and used in computing the actual dilution factor of
the samples. Observations on the methods used for NH.-N, NCL-N, and
N03-N are presented in the Appendix. COD of the sample was determined
using the rapid method (35).
Storage of Samples - All the nitrogen analyses, COD, and BOD were performed
on the samples rapidly and without storage. N02 and N03-N analyses of
samples stored with H^SO. were found to be unsatisfactory. In the deter-
mination of solids sometimes it was found to be inconvenient to process
all the samples on one day. On such occasions the samples were refrig-
erated and determinations were made as soon as possible.
Respirometry - A Gilson differential respirometer was used. All oxygen
uptake measurements were made at 20°C using 15 ml respirometric flasks.
The flasks contained a sample volume of 3 ml in the main compartment
and 0.2 ml of KOH (20% w/v) in the central well. The flasks were equili-
brated for 20 minutes before oxygen uptake measurements were started.
The total volume of the liquid in the main compartment was 3 ml in all
experiments. The effective concentration in the flask expressed in
molarity or mg/1 was computed using this sample volume. Appropriate
controls were kept in all the experimental runs. Conditions in the
flasks were run in duplicate and the results were based on an average.
Few differences in the flasks with identical substrates and conditions
were observed.
Description of Terms
a) Solids Retention Time (SRT) - This is the theoretical time that the
solids are retained in the system before leaving the system. Mathemati-
cally, the term can be represented by
CRT = # suspended solids in the system /7,x
I suspended solids leaving the system/day v '
bj Hydraulic Retention Time (HRT) - This is the theoretical time that a
particle of liquid remains in the system. HRT can be represented by
HRT _ volume of the system /72\
volume of the liquid coming into the system/day v '
In a completely mixed system operating without solids recycling, the SRT
equals the HRT.
165
-------
c) Loading Factor - The loading factor is the estimate of the weight of
organic material sent to a treatment system per weight of microbial solids
in the system. The loading factor can be expressed in any convenient
measure of the microbial oxidizability of a waste such as BOD, COD, or
TOC and of the active microorganisms such as MLVSS or TVS. In this study
the loading factor was expressed as
# COD fed to the system/day
# MLVSS in the system
.- •
d) Measures of Nitrification - To distinguish between the production of
nitrites, nitrates, and nitrites plus nitrates, the terms nitritification,
nitratification, and nitrification were used respectively. In batch
nitrification experiments, the terms were defined in the following manner:
% nitritification - This is the percent of the nitrites to the total
nitrogen in the system, i.e.,
NO,-N(mg/l) x 100
2 (74)
N02-N + N03-N + TKN(mg/l)
% nitratification - This is the percent of the nitrates to the total
nitrogen in the system, i.e.,
NO,-N(mg/l) x 100
(75)
N02-N + N03-N + TKN(mg/l)
% nitrification - This is the percent of both nitrites and nitrates to
the total nitrogen in the system, i.e.,
N02-N + N03-N(mg/l) x 100
N02-N + N03-N + TKN(mg/l) (76)
In the continuous flow studies, the percent nitritification, nitratifi-
cation, and nitrification were only related to the feed TKN in order to
indicate the quantity of TKN that was converted to the oxidized forms
of nitrogen. These terms were calculated as follows:
NO?-N(mg/l) x 100
% mtritif i cation - 2,mn/1v (77)
TKN(mg/TT
166
-------
N03-N(mg/l) x 100
% nltratlflcation TKN(mg/1) (78)
% nitrification - NVN + N03-N(.ng/1) x 100
TKN(mg/l)( '
Total nitrogen was used as an estimate of the contribution of all nitrogen
forms and was expressed as TN = N02-N + N03-N + TKN.
RESULTS
Nitrification - The nitrification studies were done in both continuous
and batch systems.
a) Continuous Flow Units - Initially a series of experiments were run in
which solids recycling was incorporated in the nitrification units. The
results of these experiments are presented in Table 18. In this and
subsequent Tables the terms % nitritification and % nitratification
were used to separate the effect of nitrite and nitrate production
respectively. Additional studies were performed with units that did
not contain solids recycling. The results of these studies are pre-
sented in Tables 19-22.
In spite of the precautions to maintain constant characteristics of the
feed suspension, the characteristics did change with time especially
when the feeds were kept in the low temperature reservoirs for a long
time. The greatest change took place in the ammonia nitrogen concen-
tration which increased anywhere between 0-15 times depending on the
time it was stored and on its initial total Kjeldahl nitrogen (TKN)
concentration. As low as 5 mg/1 and as high as 1500 mg/1 of ammonia
nitrogen (NH3-N) was noted in the feed. Depending on the strength of
each batch of feed, the concentration of soluble constituents such as
NH3-N was abruptly altered when a new batch was used. Although the same
procedures were followed in preparing feed suspensions, large amounts
of grit were occasionally encountered. The grit had a tendency to
accumulate in the aeration units and was reflected as an increase in the
total solids concentration in some of the aeration units. When this
happened, the mixed liquor total solids concentration in some runs was
greater than the total solids in the feed. In some runs this resulted
in negative COD and solids removals.
When there were no large changes in the solids retention time (SRT) of
the units, a new equilibrium could be reached quickly. However, when
the SRT was changed drastically, longer periods were needed to attain
equilibrium. Occasionally the composition of the feed batches during
a run fluctuated and the performance of some of the units was affected.
Data collected during an equilibrium period were used for evaluation.
The data in Tables 18-22 were collected during the period when the units
167
-------
TABLE 18. PERFORMANCE OF CONTINUOUS FLOW UNITS WITH SOLIDS RECYCLING:
NITRIFICATION STUDY
en
oo
Units
Days used for summary
FEED
Suspended Solids mg/1
COD mg/1
TKN mg/1
EFFLUENT
Suspended Solids mg/1
COD mg/1
TKN mg/1
NH4-N mg/1
N02-N mg/1
N03-N mg/1
pH, Aeration Unit
HRT, Total Unit, days
HRT, Aeration Unit, days
SRT, Total Unit, days
D
37-65
1730
2160
380
200
300
135
145
0
230
5.6
14
11.2
216
E
20-65
3430
3975
675
180
650
170
220
310
90
6.5
15.6
12.5
465
F
41-66
5270
6145
1150
75
916
370
200
430
70
6.5
12.4
9.9
870
continued,
-------
TABLE 18 Continued.
Units
Days used for summary
% N - Balance
% Nitritifi cation
% Nitratifi cation
% Nitrification
% COD Removal
-aeration unit
-total unit
% Suspended Solids Removal
-aeration unit
-total unit
Loading Factor
D
37-65
-4
0
60.5
60.5
-24
86
-35
88
.08
E
20-65
-16
45.9
13.3
59.2
-13
84
-56
95
.07
F
41-65
-24
37.4
6.1
43.5
19
85
-20
98
.094
-------
TABLE 19. PERFORMANCE OF CONTINUOUS FLOW UNITS: NITRIFICATION STUDY
Units
Mean SRT (daysL
Days used for summary
FEED
Total Solids mg/1
Suspended Solids mg/1
COD mg/1
TKN mg/1
UNIT (mixed liquor)
pH
Total Solids mg/1
Suspended Solids mg/1
COD mg/1
NH4-N mg/1
TKN mg/1
N02-N mg/1
N03-N mg/1
N03-N + N02-N mg/1
Total - N mg/1
% N - Balance
% Nitritifi cation
% Nitratifi cation
% Nitrification
% COD Removal
% Suspended Solids Removal
Loading Factor
A
9
35-48
1072
535
810
160
6.3
1410
1035
475
30
60
83
4
87
147
-7.5
52.2
2.5
54.7
41
-93
0.49
A1
10.6
27-93
2880
1730
2160
385
5.8
2470
1065
210
120
205
25
170
195
400
+5.2
6.5
44.0
50.5
90
39
0.26
B'
9
48-76
1300
660
865
170
5.8
928
220
365
40
75
2
68
70
145
-15.3
1.4
47.5
48.9
58
67
0.59
C
11.6
19-33
1210
595
850
160
5.6
995
390
490
45
80
4
69
73
153
+1.9
2.5
49.8
52.3
42
35
0.31
D1
12.5
52-97
2415
1310
1500
350
5.6
2070
725
1115
105
70
1
135
136
206
-12.8
0.14
38
38.1
26
45
0.19
-------
TABLE 20. SUMMARY OF THE DATA OF
Mean SRT (days)
Days used for summary
FEED
Total Solids mg/1
Suspended Solids mg/1
COD mg/1
TKN mg/1
UNIT (mixed liquor)
pH
Total Solids mg/1
Suspended Solids mg/1
COD mg/1
NH4-N mg/1
TKN mg/1
N02-N mg/1
N03-N mg/1
N03-N + N02-N mg/1
Total - N mg/1
% N - Balance
% Nitritifi cation
% Nitratifi cation
% Nitrification
% COD Removal
% Suspended Solids Removal
Loading Factor
18.6
4-67
7010
4720
5090
1030
4.85
8135
3030
3780
270
515
2
510
512
1027
-0.4
0.2
49.5
49.7
26
36
0.11
12.8
6-24
5865
2670
3515
600
6.1
3290
1820
2260
140
265
245
50
295
560
-6.4
40.8
6.8
47.6
36
32
0.19
PERFORMANCE OF UNIT E1 AT VARIOUS SRT
10.3
128-155
8200
5710
4900
850
5.8
6495
4245
4425
170
380
200
160
360
740
-16
23.8
18.9
42.7
30
26
0.15
5.1
1-14
12270
7440
8360
955
6.1
7070
4915
6090
245
485
520
30
550
1035
+8.9
54.6
3.3
57.9
27
34
0.44
3.3
20-50
7155
4655
6050
1045
6.2
9810
7655
7085
300
680
420
60
480
1160
+10.8
40.1
5.7
45.8
-17
-64
0.35
2.8
53-67
8544
5480
6205
1020
6.6
10450
7675
6755
310
620
345
225
570
1190
+16
33.8
21.9
55.7
-9
-40
0.42
1.8
17-25
12660
8025
8320
795
6.9
6805
7625
9295
250
545
330
55
385
930
+9.4
40.8
6.8
47.6
-12
5
0.84
1.0
10-28
7625
5330
6830
820
var.
11885
9885
8255
490
815
25
10
35
850
+3.5
3.3
.9
4.2
-21
-85
1.9
-------
TABLE 21. SUMMARY OF THE DATA OF PERFORMANCE OF UNIT F1 AT VARIOUS SRT
Mean SRT (days)
Days used
for
summary
13.5
140-161
12.2
7-70
5.3
12-31
2.9
53-63
1.9
17-28
0.9
19-25
tv>
FEED
Total Solids mg/1
Suspended Solids mg/1
COD mg/1
TKN mg/1
UNIT (mixed liquor)
pH
Total Solids mg/1
Suspended Solids mg/1
COD mg/1
NH4-N mg/1
TKN mg/1
N02-N mg/1
N03-N mg/1
N03-N
+ N02-N
Total - N mg/1
% N Balance
% Nitritification
% Nitratification
% Nitrification
% COD Removal
% Suspended Solids Removal
Loading Factor
10190
7480
8035
1220
7760
4820
5075
870
11960
9080
8270
1365
13365
10495
9230
1520
13130
10080
12810
1285
11520
8625
10885
1240
5.6
8440
5890
5460
285
595
210
290
500
1095
-10
17.3
23.8
41.1
32
21
0.15
6.2
11300
7865
3490
220
405
405
25
430
835
-3.7
46.7
3
49.7
31
-65
0.13
5.6
9695
6545
6655
440
855
150
430
580
1435
-5.1
10.9
31.6
42.5
20
28
0.35
6.6
17970
13815
12130
495
965
695
170
865
1830
+22.2
45.6
11.2
56.8
-32
-31
0.37
7.0
13990
9320
12670
400
910
290
135
425
1335
+4.2
22.8
10.4
33.2
1
8
1.02
var.
18695
14000
12685
765
1315
40
5
45
1360
+9.5
2.9
0.4
3.3
-17
-62
1.09
-------
TABLE 22. SUMMARY OF THE DATA OF PERFORMANCE OF UNIT G1 AT VARIOUS SRT
Mean SRT (days)
11.1
3.2
1
Days used for summary
65-147
7-31
2-32
17-24
FEED
Total Solids mg/1
Suspended Solids mg/1
COD mg/1
TKN mg/1
UNIT (mixed liquor)
PH
Total Solids mg/1
Suspended Solids mg/1
COD mg/1
NH4-N mg/1
TKN mg/1
mg/1
,-N mg/1
N02-N mg/1
Total - N mg/1
% N - Balance
% Nitritifi cation
% NiRatification
% Nitrification
% COD Removal
% Suspended Solids Removal
Loading Factor
N02-N
N03-N
12170
8305
11440
1980
6.5
11420
8905
8155
550
1005
940
65
1005
2010
+1.7
46.8
3.3
50.1
29
-7
0.2
13465
7950
10680
2155
7.0
27310
18720
14180
645
1275
965
30
995
2270
+5.4
44.8
1.4
46.2
-33
-135
0.34
17890
12960
13690
2175
7.2
20085
16650
14850
640
1250
955
40
995
2245
+9.1
46.4
1.9
48.3
-28
-8
0.72
19250
14355
18820
1745
8.2
20540
16690
19665
1100
1760
15
20
35
1795
+3.3
0.7
1.4
2.1
-16
-5
2.1
-------
performed most uniformly. Following the equilibrium period, the conditions
were changed, e.g., change of SRT, to study the performance of the units
under the new conditions.
Although the solids retention time in the continuous flow units with
recycling was very high, on the order of 216-870 days, and the loading
rate was low, only 43.5-60.5% nitrification occurred (Table 18, Figure
53) and there was a substantial amount of NH4-N present in the units.
A similar observation was made in the continuous flow units without
recycling of solids. The persistence of significant quantities of NH4~N
at reasonably long SRT values coupled with lack of complete nitrification
suggested that some factor may be inhibiting the establishment of an
adequate nitrifying population. Higher loading rates of TKN in the feed
resulted in higher NH4~N concentrations in the units (Figure 53). The
low pH of the nitrifying units as well as the repression of the synthesis
of ammonia oxidizing enzymes that might be taking place due to the accumu-
lation of NH4-N, N02-N, and NOg-N, possibly were responsible for the
incomplete oxidation of the available ammonia.
Irrespective of the concentration of ammonia in the mixed liquor of the
units, about 40 to 60% of the feed TKN was nitrified (Figure 53) when
the SRT was maintained above 1.8 days. At this SRT the nitrification
started to fall off with elimination of nitrification at an SRT of 1
day, presumably due to the wash out of the nitrifying population.
With lower feed TKN concentrations (Units A', B1, C, D'), nitrification
produced nitrates as the primary end product. With higher TKN concen-
trations (Unit G1), nitrite was the primary end product. However, in
units E1 and F1 both nitritification and nitratification occurred at
some SRT values. At other SRT values nitritification was significantly
higher than nitratification.
The formation of the type of oxidized nitrogen may depend on the nature
of the seed as well as the TKN loading. Although unit A received a
feed TKN similar to the feed TKN of the unit B1, nitrite was its primary
end product whereas N03 was the end product in the latter case. The
difference in their performance lies in the fact that unit A received
garden soil as seed whereas B1 received a highly nitrifying activated
sludge. It is likely that the garden soil was not harboring large
numbers of Nitrobacter in comparison to the Nitrosomonas or that
Nitrobacter was eliminated in the initial stages of operation of unit
A, thus resulting in the accumulation of nitrite only.
The effect of COD to MLVSS loading concentration on free NH, and undis-
sociated nitrous acid at various SRT values on the nitrification pro-
cesses is discussed later.
In the units with solids recycling, COD and suspended solids removals
were about 85% and 90% respectively. The percent removal efficiencies
174
-------
80
O 60-j
u.
a:
40-
I-
Z
LJ
CC 2C
LJ
Q.
2OO
400
MOO J
800-
600-
400-
200-
600 800
UNIT NH -N - mg/l
a WASHOUT
• UNITS WITHOUT
SOLIDS RECYCLE
o UNITS WITH
SOLIDS RECYCLE
500 1000 1500 2000 2500
FEED TKN - mg/l
FIGURE 53
TKN, AMMONIA, AND PERCENT NITRIFICATION
LABORATORY CONTINUOUS FLOW UNITS
-------
1200
0
SUSPENDED
SOLIDS
COD
100
80-
60-
40-
20-
0
10
o
z
<
•100
80
-60
-40
UJ
o
•20 £
QL
20 30
days
FIGURE 54
GENERAL RESULTS FROM NITRIFICATION
BATCH UNIT "b"
-------
400
320-
Q>
0)
Q.
240-
160-
TKN
200
FIGURE 55
NITROGEN DATA FROM NITRIFICATION
OF BATCH UNIT "b"
-------
4000-
^ 3000 J
o>
Q.
0»
2000-1
100
oo
1000 4
COD
SUSPENDED
SOLIDS
10
20
30
40 50
TIME
days
10
FIGURE 56
COD, SOLIDS, AND TOTAL NITROGEN
NITRIFICATION OF BATCH UNIT 40
-------
800-
1C
NHv.-N
160
120
80-
40-
N03-N
FIGURE 57
NITROGEN DATA FROM NITRIFICATION
OF BATCH UNIT 40
-------
given for the units with recycling represent the computations based on
the feed and mixed liquor and not on the effluent. All the units yielded
a mixed liquor containing solids that settled readily with a clear super-
natant. Analyses of the supernatant, however, were not performed, and
we are unable to present information on the absolute removal efficiencies
of COD and suspended solids.
b) Batch Studies - A study was undertaken to determine the degree of
nitrification that can be accomplished under different CODrMLVSS, and
TKN:MLVSS loadings. Suspensions of poultry manure at various concen-
trations were made as noted. A known volume of a highly nitrifying
activated sludge developed on poultry manure was added to different
volumes of poultry manure suspensions placed in eight liter aspirator
bottles. The systems were aerated with air saturated with water and
high residual dissolved oxygen (DO) levels were maintained. Analysis
of TKN, NH.-N, COD and MLVSS of the feed and seed were performed initially.
The ratios noted in the following tables for feed COD to seed MLVSS, and
the feed TKN to seed MLVSS were derived from the total weight of COD
and TKN of the feed and the MLVSS of the seed fed into the reactor on
the initial day. Routine analyses of TKN, NH4~N, N02-N, NOg-N, COD and
pH were performed on the mixed liquors after making up any evaporation
losses by adding distilled water. The mixed liquor suspended solids
were determined periodically. Figures 54-57 give the data on the per-
formance of the batch unit 'b1 and 40, which are typical of the perfor-
mance of the other units. Unit 'b1 typifies results obtained with a
dilute system whereas results of unit 40 represent those of a relatively
concentrated system. The complete operational data of all the units, i.e.,
batch units b, c, d, e, 40, 50, 60, and 70 are given in the Appendix,
Tables VI- 1-8.
The initial NO^-N + NO^-N present in the systems was due to the seed.
The poultry manure suspensions did not contain any significant amount
of nitrite and nitrate. In the units nitratification was accomplished
after active nitritification was once established.
The data on the performance of the units operated at various COD and
TKN:MLVSS loadings are presented in Table 23.
Nitrification occurred over a wide range of COD (feed): MLVSS (seed)
ratios and TKN:MLVSS ratios. These ratios were based on the MLVSS of
seed initially added and not on the MLVSS under aeration which changed
continuously because of the batch nature of the reactors. Since the
nitrifiers came only with the seed and as they were not present in
significant numbers in the poultry manure suspensions, this type of
ratio was used to estimate the food to active mass of nitrifying
organisms initially in the system. Although the COD:MLVSS, and TKN:MLVSS
ratios were very close to each other in units 'c1 and 'd1, it is inter-
esting to note that the degree of nitrification in unit 'd1 was far less
than that of 'c'. Similarly unit 'e1 did not nitrify well although its
180
-------
TABLE 23. NITRIFICATION RESULTS - BATCH STUDY
CD
Unit
b
c
d
e
40
50
60
70
* Pro
Initial feed
COD:MLVSS
in seed
12.0:1
18.1:1
19.6:1
31.8:1
40.9:1
41.9:1
68.7:1
68.7:1
.0 MH _N = ^' v
Initial feed
TKN:MLVSS
in seed
2:1
3:1
2.7:1
4.5:1
4.8:1
6.0:1
7.6:1
8.7:1
mg/1 NH4-N x 10pH
Maximum
NH4-N
measured
(tng/1 )
178
188
612
762
247
305
374
410
Maximum*
free NHg-N
in unit
(mg/1 )
23
22
168
192
32
40
54
65
Day maximum
free NH4-N
noticed
2
5
1
1
1
1
1
1
Maximum**
apparent
nitrification
55
55
17
16
65
63
59
54
pH
** The maximum nitrification value obtained during the entire study
-------
COD:MLVSS and TKN:MLVSS ratios were lower than units 40, 50, 60 and 70.
From the data it appears the critical factor that dictates the process
Of nitrification seems to be the maximum NH -N or the free NHg-N con-
centration observed during the period of operation of the units and per-
haps not the initial TKN or COD to MLVSS ratios. The maximum NH4-N
concentration in the mixed liquor resulted in most of the units within
the first two days of operation except in unit V where it resulted
after 5 days (Table 23). From the data presented in this Table, the
inhibitory concentration of total NH.-N seems to lie between 410-612 mg/1
(Figure 58). The free NhU-N concentration was 65 and 168 mg/1 respec-
tively at these concentrations of NH^-N in mixed liquor (Table 23).
A plot of the initial TKN of the suspension of feed and the maximum NH4-N
concentration measured in the units revealed approximately a 2:1 relation-
ship (Figure 58). This indicates that at least 50% of the feed TKN can
be deaminated.
Although a significant degree of nitrification was accomplished in the
units as evidenced by the formulation of nitrite and nitrates, 60-82%
Of the total nitrogen was lost from six of the eight batch reactors
(Table VI). The loss of total nitrogen was 27 and 38% in the reactors
'b1 and 'c', which had the lowest NH.-N concentration. Most of the
nitrogen loss was presumably due to the stripping of NH3-N during the
early period of operation. Although the reactors were vigorously aerated,
it was likely that anaerobic conditions prevailed within the floes and a
part of the nitrogen lost was due possibly to denitrification. This was
suggested by the decrease in the NOg-N concentration in some of the
reactors.
c] Effect of pH - In the continuous flow studies, the degree of nitrifi-
cation was found to be not more than 60% over a wide range of loadings
and detention times. In the actively nitrifying units the pH ranged
between 4.9 to 7.2. This pH range was different from the range con-
sidered to be optimum for nitrification. It was thought that the low
pH might have hindered further nitrification. To investigate this possi-
bility, a respirometric experiment was set up with a highly nitritifying
oxidation ditch mixed liquor (ODML) and a highly nitratifying mixed
liquor (unit A1). In the respirometer flasks, the pH was adjusted to
different levels (4-11) without buffering. The experimental conditions
for the respirometry have been described previously. Care was taken to
connect the flasks to the manometers immediately after the adjustment of
pH to minimize their exposure to the atmosphere. Initial analyses of all
forms of nitrogen were performed on all the samples. Oxygen uptake
measurements were continued until a reasonable plateau was obtained. At
that time, the flasks were disconnected and the final analyses of all
forms of nitrogen contained in the mixed liquor were performed to obtain
a nitrogen balance. The oxygen uptake data are presented in Figures 59
and 60 and the nitrogen data is presented in Tables 24 and 25.
182
-------
00
CO
UJ
o
(T
Ul
a:
I
801
60-
40-
20-
_ 800 1
9
I
(O
600-
:D
^ 400 -\
200-
i
0 200 400 600 800
MAXIMUM NH4-N IN UNITS - mg/l
300 600 900
INITIAL TKN IN UNITS
1200 1500
- mg/l
FIGURE 58
MAXIMUM NH4-N, NITRIFICATION AND
INITIAL TKN - BATCH NITRIFICATION UNITS
-------
10001
UNIT A'HIGH NITRATES
800 A
o»
E
600 J
LJ
o.
13
400
li)
x
o
200 J
150
FIGURE 59
CONTROL OF pH DURING NITRIFICATION
NITRATIFYING MIXED LIQUOR
184
-------
3000
2000-
liJ
*:
2
Q.
D
LU
O
X
o
1000-
OXIDATION DITCH MIXED LIQUOR
HIGH NITRITES
50 100
TIME
150
— hours
200
250
FIGURE 60
OXYGEN UPTAKE OF ODML
AT VARYING INITIAL pH VALUES
185
-------
TABLE 24. EFFECT OF pH ON THE NITRIFICATION OF UNIT A' MIXED LIQUOR
00
Flask
PH
4
5.7
6
7
8
9
10
11
TKN
Initial Final
- (mg/1)
530
539
530
528
530
528
530
524
514
420
434
285
317
428
477
449
NH4-N
Initial Final
(mg/1 )
320
317
320
317
320
311
320
326
343
280
261
140
144
171
209
120
Ini
2
2
2
2
2
2
2
2
N02-N
tial Final
(mg/1)
.4
.5
.4
.4
.4
.4
.4
.3
0.8
1.0
1.8
4.2
1.0
3.2
2.2
3.2
N03-N
Initial Final
(mg/1 )
523
478
523
533
523
509
523
560
612
549
572
590
599
558
580
615
Total N
Initial Final
(mg/1 )
1055.4 1126
1019.5 970
1055.4 1008
1063.4 879
1055.4 917
1039.4 989
1055.4 1059
1086.3 1067
Nitrogen
Balance
(% change)
+6.7
-4.9
-4.5
-16.9
-13.1
-6.1
+0.4
-1.8
Initial Day: March 9, 1971
Final Day: March 19, 1971
-------
TABLE 25. EFFECT OF pH ON THE NITRIFICATION OF OXIDATION DITCH MIXED LIQUOR
co
Flask
Initial
4
5
6
7
8
9
10
11
pH
Final
5.3
6.3
7.5
8.4
8.2
8.1
9.5
8.7
TKN
Initial
(mg/1
1790
1768
1786
1768
1768
1768
1768
1768
Final
)
1928
1921
1877
1663
1589
1592
1589
1598
NH4-N
Initial Final
(mg/1 )
353
353
353
353
348
353
360
348
408
368
350
297
294
547
150
534
N02-N
Initial Final
(mg/1 )
1172 692
1172 1167
1172 890
1172 742
1172 818
1172 1457
1172 13
1172 1629
N03-N
Initial Final
(mg/1 )
254
249
253
249
249
257
.4 249
240
454
326
322
296
300
320
49
53
Total N
Initial Final
(mg/1)
3207
3189
3211
3189
3189
3197
3189
3180
3074
3414
3089
2701
2707
3369
1651
3280
Nitrogen
Balance
(% change)
-4
+7
-3.8
-15
-15
+5.3
-48.2
+3.1
Initial Day: March 9, 1971
Final Day: March 19, 1971
-------
The oxygen uptake data of the ODML revealed that except for the system
initially adjusted to pH 4, all the flasks were active, although the
pH 10 and pH 11 flasks showed an initial lag of 40 and 70 hours respec-
tively. The flasks adjusted to pH 7, 8, and 9 exerted higher uptakes
at a much faster rate than the control (pH 6.1) but there did not seem
to be any significant differences in their cumulative oxygen uptake. It
is interesting that although the pH 11 flask exhibited a long lag ini-
tially, it exerted the highest cumulative oxygen uptake among all the
flasks. The higher exertion of oxygen uptake was perhaps due to the
readily assimilable low molecular weight compounds that may have been
formed because of the hydrolysis of complex substances at pH 11. A
similar phenomenon has been reported with municipal wastes (185). The
long initial lag could be due to the high initial pH which is unfavorable
for microbial activity. The initial pH of 11 decreased to 8.7 by the
end of the run. The end of the lag period may have occurred as the pH
decreased to more favorable levels.
The nitrogen data for the pH 11 flask (Table 25) indicated that nitrifi-
cation took place. This was observed by the increase in the nitrite
content by the end of the run. A part of this increase could be due
to the reduction of nitrate. The ammonia concentration increased from
348 to 534 mg/1 in this flask. Even at this concentration of NH^-N, nitri-
fication took place. It is likely that the nitrification started to occur
when the pH dropped in the flask and a rapid rate of oxygen uptake occurred,
In contrast to the pH 11 flask, the pH 10 flask lost a high amount of
nitrogen (about 48%) in spite of significant oxygen uptake. It was
observed that the mixed liquor in the flask became gelatinous and it is
possible that oxygen transfer was inhibited rendering the mixed liquor
anaerobic and causing denitrification.
In pH 7 and 8 flasks, although an apparent increase in N03 was observed,
there was loss of nitrogen although not as high as in the pH 10 flask.
The pH 9 flask exerted similar rates of oxygen uptake as the pH 7 and
8 flasks but it nitrified and there was no significant loss in nitrogen.
In the pH 4 flask, there was no significant oxygen uptake although there
was an increase in nitrates and decrease in nitrites. This may be due
to a chemical reaction (Equation 62) rather than a microbiological one.
Although the flasks were shaken vigorously to keep the conditions aerobic,
it was possible that reducing conditions prevailed at the floccular level
in most flasks. If this condition occurred, denitrification was favored.
Denitrification of nitrates has been reported in aerobic respirometric
studies. Considering the soupy nature of this mixed liquor, the slightly
positive nitrogen balances might be due to sampling error in performing
the analysis.
With the highly nitratifying wastes (unit A1), adjustment of the pH of
the mixed liquor initially to pH 6, 7 and 8 resulted in a more rapid
188
-------
oxygen uptake rate and a higher cumulative oxygen uptake (Figure 59).
There did not seem to be any significant difference between the oxygen
uptake of the control and the mixed liquor adjusted to pH 4. pH 10
and 11 were inhibitory and the patterns of oxygen uptake were not the
same as those for ODML (Figure 60). There were some apparent increases
in the nitrates in all the flasks but the increases were not signifi-
cant or able to be attributed to nitrification.
From these studies on the mixed liquors of a highly nitritifying and
highly nitratifying system treating poultry wastes, it does not seem
to be advantageous to adjust the pH of a system to enhance nitrification.
The increased exertion of oxygen uptake observed in the pH range of 6-8
(unit A1) and 7-9 (ODML) was likely due to increased carbonaceous demand
since no significant increases in the oxidized forms of nitrogen were
noted.
d) Effect of Dilution - Certain toxic industrial wastes are known to
exert a type of BOD known as the "sliding scale" BOD. Such wastes exert
a greater BOD at higher dilutions than at lower dilutions. At the
higher dilutions the toxicity of the waste is decreased and the micro-
organisms find an environment conducive for their growth. Observations
on the determination of the BOD of a concentrated poultry waste mixed
liquor indicated that a greater BOD was exerted at higher dilutions.
Subsequently, in an empirical respirometric study, these observations
were verified and it was established that a 1:2 dilution (mixed liquor:
water) appeared optimum. To determine the effect of dilution on the
oxygen uptake of an oxidation ditch mixed liquor (ODML) containing high
solids concentration (1.9%), a detailed respirometric experiment was
performed. In this study, some of the 1:2 dilution flasks were supple-
mented with combinations of NH.-N, P^-N, and N03-N to adjust the con-
centration of these compounds to concentrations comparable to those in
the undiluted flask. In this way the effect of nitrogen concentrations
on oxygen uptake and nitrification patterns of dilute systems was
evaluated.
Nitrogen analyses were performed in all the respirometric flasks before
and at the end of the run. A nitrogen balance was made to find out
whether the dilution and supplementation with various forms of nitrogen
had any effect on nitrification.
To compare the oxygen uptake of the diluted suspensions with the undiluted
mixed liquor, the oxygen uptakes in the flasks containing the diluted
suspensions were multiplied by the appropriate dilution factors. This
adjustment placed the oxygen uptake data in the various flasks on a
comparable basis, essentially oxygen uptake per solids concentration
in the control.
Figure 61 represents the adjusted oxygen uptake of the various flasks
and the actual nitrogen data in the flasks are presented in Table 26.
The respirometric data indicated that the oxygen uptake exerted at the
189
-------
o>
2500
2000
I5OO-
Ul
1000-
Z
UJ
e>
*
8 5OO-
o
LJ
H-
tn
o
DILUTION
25O01
20QO-\
1500-
1000-
500-
DILUTION a SUPPLEMENTATION
IOO 150 0 50
TIME - hours
ODML 12 WITH
NOg-N, N03-N,
NH4-N
OOML 1=2 WITH
NOg-N, N03-N
100
150
200
250
FIGURE 61
OXYGEN UPTAKE OF DILUTED MIXED LIQUOR
AND MIXED LIQUOR WITH ADDED NITROGEN -
NITRIFICATION STUDY
-------
TABLE 26.
ODML
Dilution*
Undiluted
(control )
1:1
1:2
1:3
1:2
NITROGEN DATA
RESPIROMETRIC
TKN, mg/1
Initial
930
466
294
238
284
OF ODML AT
STUDY
NH4-N, mg/1
Initial
140
70
42
33
42
VARIOUS
N02-N
Initial
507
240
160
126
357
DILUTIONS
1, mg/1
Final
539
260
205
130
464
-
N03-N,
Initial
175
88
58
44
170
mg/1
Final
318
190
150
90
224
(with N02+N03-N)
1:2
(with NH4+N02+N03-N)
434
135
350
445
155
248
* Ratio of mixed liquor to water, volume basis.
TABLE 27. ADJUSTED NITROGEN
Sample
Dilution**
Undiluted
1:1
1:2
1:3
N02-N, mg/1
Initial Final
507
480
480
504
539
520
615
520
DATA*
N03-N, mg/1
Initial Final
175
176
174
176
318
380
450
360
**
Based upon data in Table 26 and the dilutions.
Ratio of mixed liquor to water, volume basis.
191
-------
1:2 dilution was much higher than the undiluted or 1:1 mixed liquors.
Although the oxygen uptake of the 1:3 flask was higher than the control
and 1:1 flask, it was lower than the one for the 1:2 flask. This dif-
ference between the uptake rates in the 1:2 and 1:3 flasks may have
been due to higher numbers of nitrifying and other organisms present
in the 1:2 flask.
A comparison of the oxygen uptake patterns of 1:2 diluted mixed liquor
and the 1:2 diluted mixed liquor supplemented with nitrogen (Figure 61)
showed that the oxygen uptake was lowered even though nitrification
occurred. The inhibitions appeared to be due to NC^-N and N03-N since
there was no appreciable difference in oxygen uptake when 135 mg NH^-N/1
was in the flasks.
The initial and final N02-N and N03-N values of the various dilutions
were adjusted to levels comparable to that of the undiluted mixed liquor
by multiplying with the appropriate dilution factors (Table 27). The
nitrification also increased up to the 1:2 dilution and decreased at the
1:3 dilution. Perhaps the adjusted oxygen uptake and the degree of
nitrification of the 1:3 flask would have been comparable to that of
the 1:2 flask if the period of experimentation was prolonged to increase
the population of nitrifiers and other organisms.
As indicated earlier, the lowered nitrification in the 1:3 flask may
have been due to inadequate numbers of nitrifying bacteria.
The studies indicated that oxygen uptake and the degree of nitrification
increased as a concentrated mixed liquor was diluted. The data suggest
that a solids concentration of 1% or less favors higher uptake rates
and nitrification. Dilutions of a nitrifying mixed liquor will decrease
the population of nitrifying bacteria. Beyond a certain dilution this
is reflected as a decrease in the rate of nitrification. With the opti-
mum dilution found in the study, increases in the NCL-N and NCL-N coiir
centrations decreased oxygen uptake rates although some degree of nitri-
fication was obtained.
The denitrification of the mixed liquor used in this study was also eval-
uated at various dilutions and the implications were discussed elsewhere.
e) Effect of NO^-N and NO,-N Concentration - Pure culture studies have
shown that high~~concentrations of nitrite are inhibitory to nitrification,
particularly at lower pH ranges. The oxidation ditch at the Cornell
Waste Management Laboratory treating poultry manure routinely contained
high concentrations of nitrites and lesser concentrations of nitrates.
If aeration of poultry manure becomes more common, significant concen-
trations of nitrites and nitrates may occur in such systems.
To investigate whether high concentrations of N02-N are inhibitory to its
further oxidation to nitrate, a respirometric study was undertaken. The
192
-------
suspended solids in each respirometric flask was maintained at the same
level although the N02-N and N03-N concentrations were varied in the
following manner. A known amount of distilled water was added to a
given volume of oxidation ditch mixed liquor and centrifuged for 5 minutes
at 20,000 rpm. A volume equal to the amount of distilled water added was
withdrawn carefully from the supernatant without disturbing the solids
pellet at the bottom. By manipulating the volume of distilled water
added, and removing the same volume of supernatant after centrifuging,
it was possible to get any degree of dilution of the N02-N and N03-N
contained in the mixed liquor, keeping the suspended solids concentration
constant.
No supplementation of nitrogen was used in this study. Unlike the pre-
vious study, the solids concentration was constant and the N02-N and
N03-N concentration was varied. Because the solids concentration was
constant, no adjustment of the oxygen uptake data was made. The actual
uptake data are reported in Figure 62.
The following dilutions were used:
Flask No. 1) Control; as is oxidation ditch mixed liquor (ODML)
2) 50 ml of ODML + 30 ml D.W. - 30 ml of supernatant
3) 50 ml of ODML + 100 ml D.W. - 100 ml of supernatant
4) 25 ml of ODML + 125 ml D.W. - 125 ml of supernatant
5) 12 ml of ODML + 120 ml D.W. - 120 ml of supernatant
6) 6 ml of ODML + 120 ml D.W. - 120 ml supernatant
The nitrogen data is presented in Table 28.
Although it is very difficult to draw any clear cut conclusions from the
respirometric data with respect to the effect of nitrite and nitrate
concentration on oxygen uptake, the trends in the oxygen uptake patterns
tend to indicate that the most concentrated systems, undiluted and 5:3,
exerted slower rates of oxygen uptake than the rest. Surprisingly, these
were the only two flasks in which a net nitrification resulted without
any significant nitrogen losses. Although the remainder of the flasks
exerted a relatively high cumulative oxygen uptake, significant nitrogen
losses occurred, an observation similar to the one made in batch nitri-
fication studies with high rates of aeration.
To observe the effects of N03-N on the oxygen uptake of a highly nitrati-
fying mixed liquor, a respirometric study was undertaken with unit A1
mixed liquor along the lines described above. The results are presented
in Figure 63 and Table 29.
The 02 uptake data did not indicate any significant differences between
the high and low nitrate flasks suggesting that effect of nitrate was
193
-------
2200
2000-
o>
1500-
LU
*:
?
Q.
1000
•z.
llJ
500
• CONTROL ML
o 5=3 ML'DW
a 1=2 ML DW
O 1-5 ML'DW
• I'10 ML'DW
• I>20ML=DW
AVERAGE MLSS
IN EACH FLASK- 20,420 mg/l
50 100
TIME
150
hours
200
250
FIGURE 62
OXYGEN UPTAKE OF DILUTED MIXED LIQUOR
NITRIFICATION STUDY
194
-------
en
TABLE 28. NITROGEN BALANCE IN A HIGHLY NITRITIFYING OXIDATION DITCH MIXED LIQUOR -
RESPIROMETRIC STUDY
Sample
Dilution*
Undiluted
5:3
1:2
1:5
1:10
1:20
TKN (mg/1)
Initial Final
1365
1291
1285
1265
1255
1241
1457
1344
980
868
812
840
NH4-N
Initial
283
184
106
70
25
15
(mg/1)
Final
364
84
56
14
14
14
N02-N 1
Initial
926
560
349
210
85
44
Final
730
283
5
2
4
1
N03-N 1
Initial
227
101
67
47
20
11
Final
385
203
68
14
11
7
TN (mg/1)
Initial Final
2518
1952
1634
1522
1360
1296
2572
1830
1053
884
837
848
Ratio of mixed liquor to water, volume basis
-------
400
CONTROL ML
5'3 ML 2 ML'DW
DW
AVERAGE MLSS IN EACH
FLASK - 2330mg/l
50 100
TIME
150
200
250
hours
FIGURE 63
OXYGEN UPTAKE OF DILUTED
NITRIFYING MIXED LIQUORS
196
-------
10
-si
TABLE 29. NITROGEN BALANCE IN THE HIGHLY NITRATIFYING UNIT A1 MIXED LIQUOR -
RESPIROMETRIC STUDY
Sample
Dilution*
Undiluted
5:3
1:2
1:5
TKN (mg/1)
Initial Final
455
336
266
182
462
350
266
196
NH4-N
Initial
231
146
84
39
(mg/1 )
Final
280
186
105
56
N02-N 1
Initial
1.1
1.0
0.6
0.4
[mg/1)
Final
.4
.4
.2
.2
N03-N i
Initial
517
225
144
77
Final
580
310
182
109
TN (mg/1)
Initial Final
973
562
411
259
1042
660
448
305
*
Ratio of mixed liquor to water, volume basis.
-------
minimal, if any, on the respiration of the mixed liquor population. The
decreased oxygen uptake in the 1:5 flask may have been due to bacteria
and growth factors removed in the supernatant. Unlike in the case of
ODML, there were no significant losses of nitrogen from any of the flasks.
This may have been due to the lower MLSS concentration which minimized
the loss of nitrogen by local denitrification. Nitrification occurred
in the system. It was not proportional to the concentration of initial
nitrate in the system suggesting that within the initial N03-N concen-
trations used, nitrification was not significantly affected.
These studies indicated that nitrate plus nitrite nitrogen concentrations
less than about 500 mg/1 had no effect on the intrinsic oxygen uptake
patterns. Combined concentrations higher than 660 mg/1 appeared to
retard the intrinsic oxygen uptake. Even with the high oxygen uptake
rates and nitrification observed, there was still residual NH4-N in
the flasks indicating that complete nitrification was not achieved.
f) Effect of NH.-N Concentration - Ammonium chloride at 0.1 M concen-
tration was reported to inhibit nitrification in domestic sewage (114).
To correct for nitrification in the BOD of the mixed liquor from one of
the units in this study that contained a relatively high concentration
NH4-N, we tried to inhibit the nitrification by addition of a 0.1 M NH4C1
solution and found that nitrification was not inhibited. On the contrary,
the NH»C1 treated samples exerted a higher BOD than the untreated samples.
To indicate the magnitude of the differences, results from one experiment
are given below.
TABLE 30. BOD OF UNTREATED AND 0.1 M NH.Cl TREATED
POULTRY MANURE MIXED LIQUOR 4
BIOCHEMICAL OXYGEN DEMAND - mg/1
SAMPLE UNTREATED
A 490
B 1540
C 2180
D 1210
E 2300
TREATED WITH 0.1
1460
2420
3840
1930
3700
M NH.C1
~ 4 —
Similar responses to 0.1 M NH4C1 were observed in other runs. The higher
BOD exerted by the 0.1 M NH4C1 treated samples presumably was due to the
utilization of NH4C1 as a substrate by the nitrifying organisms suggesting
that the mixed liquor was well acclimated to high concentrations of NH4-N.
198
-------
An oxygen uptake study was conducted to investigate the rates of nitri-
fication at different concentrations of NH.-N. The mixed liquor from
unit G1, operating at 11 days SRT, (Table 22) was used for these experi-
ments. Earlier experiments indicated that oxygen uptake patterns in a
1:2 dilution of poultry waste mixed liquor were the highest (Figure 61).
Therefore this dilution was used to study the effect of ammonia concen-
tration. Initial ammonium chloride concentrations of 0.1 to 0.5 M
(1401-7005 mg N/l) in respective flasks were obtained. These flasks and
a control were used at the 1:2 dilution. To again compare the effect
of dilution, an undiluted mixed liquor sample was included in the study.
The oxygen uptake data was multiplied by the dilution factor to place
them on a comparative basis.
An oxygen balance was made to separate the effect of carbonaceous and
nitrogenous oxygen demand. The theoretical nitrogenous oxygen demand
was computed by multiplying the NO^-N and NO--N increases by 48/14 and
64/14 respectively. The computed total nitrogenous oxygen demand was
subtracted from the observed final cumulative oxygen uptake to obtain
the carbonaceous oxygen demand (Table 31).
The oxygen uptake data (Figure 64) indicated that ammonium chloride did
not inhibit nitrification completely. The cumulative oxygen uptake for
the diluted control plus 0.1, 0.2, and 0.4M NH^Cl flasks decreased pro-
gressively; nevertheless, nitrification did take place in these systems
as indicated by the increase in the oxidized nitrogen (Table 31). The
0.5M NH.C1 flask showed adaptation and the final cumulative oxygen up-
take of this system was comparable to that of undiluted control.
Accurate comparison of the performance of the various flasks was not
possible because the oxygen uptake and analytical data of the controls
and 0.5M NH.C1 flask were based on a 7 day period whereas the data for
the remainder of the flasks were based on a 5 day period. Nevertheless,
certain conclusions can be made from this study. The cumulative oxygen
uptake of the diluted control was higher than the undiluted control
confirming the observations made previously on the effect of dilution
on oxygen uptake and nitrification (Tables 26-27). The degree of nitri-
fication was also higher in the diluted mixed liquor than the undiluted
mixed liquor as indicated by the increase in I^-N and NOj-N.
The theoretical nitrogenous oxygen demand was comparable numerically for
the diluted control and the 0.1, 0.2M NH4Cl-treated flasks (Table 31)
and it was not proportional to the amount of NH.C1 added. This suggests
that the rate of nitrification decreased with an increase in the NH4-N
concentration. Nevertheless, the nitrification did not stop even at the
0.5M NH.C1 concentration. A decrease in the carbonaceous demand was
observed at the 0.5M level with no appreciable decrease at lower
levels of NH4C1 concentration.
199
-------
TABLE 31. EFFECT OF NH4-N ON THE NITRIFICATION OF UNIT G1 MIXED LIQUOR
RESPIROMETRIC STUDIES - UNADJUSTED UPTAKE DATA
Flask
Control ,
undiluted*
Control , diluted*
(1:2 MLrdistilled water)
IS3
o
0 1:2 + 0.1M
NH4C1
1:2 + 0.2M
NH4C1
1:2 + 0.4M
NH4C1
1:2 + 0.5M*
NH4C1
Observed
increase
in N02-N
(mg/1 )
63
50
57
30
-24
29
Observed
increase
in N03-N
(mg/1 )
3
9
7
20
28
10
Theoretical Op
uptake of observed
N02-N plus N03-N
increase (mg/1)
230
211
227
194
32
145
Cumulative
02 uptake
observed
(mg/1 )
1080
492
492
456
402
356
Oxygen uptake
due to
carbonaceous
demand (mg/1 )
850
281
265
262
370
211
* based on the 1st and 7th day analysis: data for the remainder of the samples are based on the 1st
and 5th day analysis
-------
I50O
ro
o
I50O-
50
100
1000-
500
/
/
UNADJUSTED
150 200 O 50
TIME - hours
100
ISO
200
FIGURE 64
EFFECT OF AMMONIA ON THE OXYGEN UPTAKE
UNIT G' NITRIFICATION STUDY
-------
To determine the effect of the added NH4-N on the nitrification of the
various flasks, a plot of the NH4-N concentration in the flasks and the
resultant % of nitrification was made (Figure 65). The % of nitrifi-
cation in this case was defined as
Increase in N02-N + N03-N x 100 ^
initial TKN
Figure 65 suggests that the % nitrification decreased with an increased
concentration of NH»-N. The maximum nitrification of 15% accomplished
with the diluted control reflects the additional degree of nitrification
possible upon dilution of the already nitrified mixed liquor of unit G1.
This additional degree of nitrification was only about 6% when the unit
G1 mixed liquor was not diluted. This suggests that this decrease in
the degree of nitrification was presumably due to the higher concentra-
tions of NH4-N (615 mg/1), N02-N, NOg-N and other factors in the undi-
luted control than in the diluted control.
To determine whether similar oxygen uptake patterns would occur with
other mixed liquors developed on lower concentrations of ammonia, a
respirometric study using mixed liquor of units E1 and F1 (5 day SRT)
was undertaken. These units contained 312 and 492 mg of NO/1 respec-
tively as compared to the 615 mg NH.-N/l contained in the unit G1 mixed
liquor. In this study flasks containing diluted mixed liquor (1:2 ML:water)
were treated with NH^Cl to obtain 0.1-0.5M concentrations in the flasks.
Figure 66 shows the results obtained.
The oxygen uptake of all the ammonium chloride treated flasks in the case
of unit E1 mixed liquors was less than oxygen uptake of their control.
In contrast, the diluted mixed liquor of unit F1 treated with 0.1M NH.C1
showed a higher oxygen uptake than its control. The 0.2, 0.4, and 0.5M
NH4C1 treated flasks showed considerably less oxygen uptake than that of
their control. The oxygen uptake patterns of the unit E1, unit F1, and
unit G1 mixed liquors (Figures 64 and 66), with their respective controls,
showed that the unit G1 mixed liquors had the highest tolerence to NH.C1
followed by F1 mixed liquor. This was indicated by the fact that the
unit G1 mixed liquor exerted significantly higher oxygen uptake with
0.1 and 0.2M NH4C1 than its control and that with the unit F1 mixed
liquor only the 0.1M NH4C1 treated flask exhibited higher oxygen uptake
than its control.
These differences in behavior towards NH4C1 can be due to the adaptation
of the microbial population functioning in these units to varied amounts
202
-------
y
u_
a:
UJ
o
cc
UJ
CL
20
5-
co I
» NH4-N
o TKN
I DILUTED CONTROL {I = 2 M L' D W)
2 COMTROLj 'AS IS' MIXED LIQUOR
SI5mg/l NH4-N IN CONTROL
DILUTED CONTROL & O.I M KH4CI
DILUTED CONTROL a 0.2 M NH4CI
DILUTED CONTROL a 0.4M NH4CI
DILUTED CONTROL a 0.5 M NH4CI
''^XTtf&^v^ff* &
4000 8000
TKN, NH4-N -mg/l
FIGURE 65
PERCENT NITRIFICATION RELATED
TO AMMONIA AND TKN CONCENTRATIONS
203
-------
1400
14001
IOOO-
500
.£. 0.2 M
O.IM
ML'OW
CONTROL
hours
FIGURE 66
EFFECT OF NH4C1 CONCENTRATIONS
ON THE OXYGEN UPTAKE OF MIXED LIQUOR
FROM UNITS E1 AND F1
-------
of NH^-N contained in the nitrifying mixed liquors. The unit G1 mixed
liquor which contained 615 mg of NH^-N/l had the highest tolerance to
NH4-N and exerted higher oxygen uptakes with 0.1 and 0.2M NH.C1. The
unit F1 mixed liquor had less tolerance to NH.-N than unit G1 and
exerted a higher oxygen uptake with 0.1M NH.C1 than its control. In
unit F', the NH^-N concentration of the mixed liquor was 492 mg of NH.-N/l
which was less than the NH^-N concentration of unit G' mixed liquor. The
unit E1 mixed liquor had the least tolerance towards NH.-N and the oxygen
uptake exerted by all the NH^Cl treated flasks were lower than the control
The unit E1 mixed liquor contained only 312 mg of NH.-N/l. Thus the
differences in the behavior of the mixed liquors is presumably due to the
adaptation of the mixed population to varied amounts of NH.-N contained
in the nitrifying mixed liquor.
To find out whether the tolerance of the mixed liquors for high concen-
trations of NH.-N is related to the SRT, i.e., the time the micro-
organisms are exposed to the NH.-N in the units, a respirometric study
was undertaken. Mixed liquors from units E1, F1, and G1 operated at
about 3 days SRT were used. Aliquots of the mixed liquor were mixed
with NH.C1 to produce a 0.1M NH.C1 concentration in the flasks. Along
with appropriate control flasks, the oxygen uptake was determined. The
results obtained in this experiment (Figure 67) were compared with results
obtained in the previous study from mixed liquors at higher SRT values
(Figures 64 and 66).
The cumulative oxygen uptake of unit G1 mixed liquor treated with NH^Cl
was slightly higher than the oxygen uptake of the control, whereas the
oxygen uptake of the unit E1 mixed liquor treated with NH^Cl was signifi-
cantly lower than its control. Although both the untreated and NH4C1
treated mixed liquor of unit F1 exhibited no significant difference in
the oxygen uptake pattern up to 35 hr, the longer cumulative oxygen
uptake for the NH.C1 treated system was lower than that of the control.
A comparison of the oxygen patterns of the mixed liquors developed at
various SRT values (Figure 64, 66, 67) indicated that mixed liquors of
units operating at all SRT values and NH.-N concentration tended to
develop a significant tolerance to high NH4~N concentrations (Table 32).
In biological waste treatment it is known that to maintain optimum per-
formance of a system, the microbial population has to be maintained in
an active physiological state. To find out whether the tolerance for
high concentrations of NH4C1 is preserved by microbial population even
205
-------
1800
1600-1
1600
1200
800
400
1800
UNIT F MIXED LIQUOR
a O.I M NH4CI
FIGURE 67
EFFECT OF AMMONIA ON THE OXYGEN
UPTAKE OF NITRIFYING MIXED LIQUORS
206
-------
TABLE 32. EFFECT OF SRT ON THE TOLERENCE OF
AMMONIUM CHLORIDE BY MIXED LIQUORS
0, uptake (0.1M NH.Cl)
-N in 2 _ 4
4 nitrifying 0? uptake (control)
the unit units *
Unit (mg/1) (days) 40 hr. 80 hr. 120 hr.
E' 245 5.1 0.82 0.93 0.96
F1 440 5.3 1.52 1.07 1.05
G' 550 11.1 1.01 1.11 1.12
E1
F1
G1
300
495
645
3.3
2.9
3.2
0.74
0.96
1.02
0.71
0.87
1.08
0.69
0.84
1.09
under altered environmental conditions, one day old samples of mixed liqupr
from units E1, F1, and G1, about 3 days SRT, were stored at 20°C in open
beakers and subsequently used with 0.1M NH.C1. Oxygen uptake was measured
and compared with oxygen uptake of their respective controls in a respiro-
metric study. The oxygen uptake patterns of these mixed liquors are given
in Figure 68.
There does not seem to be a significant difference between the cumulative
oxygen uptake of the 0.1M NH4C1 treated flasks and control flasks. All
flasks containing NH.C1 exhibited less oxygen uptake than their respec-
tive controls including unit G1 mixed liquor which in all previous runs
had exhibited significantly higher oxygen uptake with NH.C1. A com-
parison of the previous oxygen uptake patterns of this study and those
in Figure 67 suggest that removing the microbial population from an
actively aerating system and keeping it under quiescent, unaerated con-
ditions, even temporarily, might reduce its ability to withstand higher
concentrations of ammonia when it is exposed once again to a highly
aerobic environment. Further investigations are needed to find whether
such a population will regain its ability to adapt to higher concentra-
tions of ammonia.
These studies indicated that nitrification in an actively nitrifying
poultry waste is not inhibited by 0.1M NH4C1. Previous studies with
sewage indicated that this concentration would inhibit nitrification.
Higher NH. concentrations tended to decrease the degree of nitrification.
207
-------
IZ001
800 J
400 4
UNIT £' MIXED ^
LIQUOR
t MIXED LIQUOR
a o.i M NH4ci
40
TIME
80
hours
120
1600
1200
800
X 400-1
o
UNIT 6' MIXED LIQUOR-
40
TIME
UNIT G' MIXED LIQUOR
a 0.1 M
80
hours
120
1800
1600-1
1400
1200
UJ
II
Q.
800
X 600
O
400
20O
UNIT F'
MIXED LIQUOR
UNIT F' MIXED LIQUOR
8 O.I M NH4CI
40
TIME
80
hours
I29
FIGURE 68
EFFECT OF AMMONIA ON THE OXYGEN
UPTAKE OF STORED NITRIFYING MIXED LIQUORS
208
-------
Nevertheless, the microbial population of the highly nitrifying mixed
liquors were found to be very tolerant of high concentrations of NH4-N
as indicated by the occurrence of nitrification even at 0.5M NH.C1. This
tolerance appeared related to the NH4-N concentrations to which these
populations were exposed, and perhaps to the physiological state of the
organisms.
g) Nature of Nitrification - Nitrification caused by heterotrophic
organisms is known to occur in soils and other ecosystems such as rivers,
streams, and sewage. To find whether heterotrophic nitrification was
taking place in a highly organic environment such as aerobic poultry
wastes, a study was undertaken. N-serve (2-chloro-6-trichloro-methyl-
pyridine) was used as a differential inhibitor. This compound was
reported to inhibit autotrophic nitrification, particularly the oxidation
of ammonia to nitrite, without significantly affecting the oxidation of
nitrite to nitrate (112).
A nitrifying mixed liquor (250 ml) and 27.5 mg of N-serve were initially
added to one batch reactor containing 2.5 liters of a poultry manure
suspension (1450 mg/1 of suspended solids). Periodically 13.8 mg of
N-serve solution was added to this reactor to overcome losses of the
chemical. Another batch reactor was prepared in a similar way but with-
out N-serve and used as a control. Routine analysis of COD, TKN, NH.-N,
NOp-N, and NO^-N, pH, and periodic analysis of suspended solids were
made on the mixed liquor of these two reactors.
The results (Figure 69) indicate that there was no increase in the
NOp-N or N03-N in the reactor treated with N-serve whereas a signifi-
cant amount of oxidized nitrogen was formed in the control reactor. The
pH of the controlled reactor at the end of the experiment was signifi-
cantly lower than the unit treated with N-serve.
It is evident that the formation of nitrite or nitrate in the poultry
waste was not caused by heterotrophic organisms since N-serve did inhibit
nitrification.
Denitrification - In order to gain an understanding of the various factors
that govern the denitrification of the nitrified chicken manure, several
batch experiments were performed and the results of each experiment are
presented. The denitrification protocol described under the section
Materials and Methods was used in all runs.
a) Run_ I_ - To initiate investigations of the efficiency of a batch deni-
trifying reactor, nitrifying mixed liquor from unit A1 containing 2680
mg/1 of total solids was subjected to denitrification. About 35% of
the N03-N and N02-N was lost during the first six hours without any
further appreciable loss (Figure 70). The rate of denitrification
209
-------
1201
0
0
r!2
CONTROL
CONTROL + N-SERVE
PH
•o-o.
COiMTROL
PH
CONTROL+N-SERVE A
/ N02*N03-N
\A—A
10
TIME
15 20
- days
FIGURE 69
INHIBITION OF NITRIFICATION
WITH N-SERVE
8
•4
25
I
0.
210
-------
X.
D>
E
250
200-
150-
I
ro
O
100-
i
CVJ
O
TIME
hours
FIGURE 70
DENITRIFICATION OF UNIT A'
MIXED LIQUOR
RUN I
211
-------
observed in this run was approximately 4.5 mg of N/g of total solids/
hour, within the first 6 hour period.
b) Run II - To study the effect of temperature, a nitrifying mixed liquor
containing 2280 mg/1 of total solids was taken into two reactors and
denitrification was carried out at 20°C and 35°C. From the results
presented in Figure 71, the rate and degree of denitrification was
higher at 35°C. Nevertheless complete denitrification could not be
accomplished in the system. The nitrogen losses were 5.7 and 7.4 of
N/g of T.S./hour at 20°C and 35°C respectively in the initial hours.
Denitrification occurred in the first three hours without any significant
losses afterwards.
The slight increase in N02-N may have resulted from the reduction of
nitrate. No changes in the TKN or NH4-N were observed at either tempera-
ture. COD and BOD reaction rates were higher at the higher temperature.
c) Run, IH_ - In the previous runs, the inherent hydrogen donating capa-
city, i.e., the oxygen demand of the endogenous reserves of the mixed
liquor, was utilized to bring about the denitrification. One of the
reasons for the incomplete denitrification in these systems was perhaps
due to the insufficient hydrogen donating ability of the system. To
study this possibility, an experiment was set up in which a poultry
manure suspension was added to the denitrification unit to supplement
the hydrogen donors already present.
The mixed liquor COD and SS of 1450 mg/1 and 1100 mg/1 was increased to
2580 and 1850 mg/1, respectively, by adding a freshly prepared chicken
manure suspension. Two denitrification reactors were set up with this
feed and mixed liquor suspension and the progression of denitrification
was studied at 20°C and 35°C for a longer period than before.
As in the previous runs, denitrification of the mixed liquor started
immediately and there was a rapid decrease in the NOp-N and N03-N for
about 6 hours in both reactors (Figure 72). After about 6 hours, there
was no further decrease in the N03-N at 20°C up to about 24 hours, after
which rapid denitrification once more set in. Although a slight change
in the rate of denitrification was observed after 6 hours in the reactor
at 35°C, the denitrification proceeded without any significant plateau.
The TKN, BOD, and COD removals followed trends similar to those observed
in Run II. However, the NH4-N concentrations increased considerably at
both temperatures due to the degradation of the added poultry waste. The
addition of chicken manure resulted in a higher percentage removal of
N03 plus N02-N, 87 and 70%, than observed in previous runs. These
removals correspond to an overall removal rate of 1.5 and 0.86 mg of
N/g of SS/hour at 35 and 20°C respectively over the length of the exper-
iments. The lower rates of removal observed in this run were due to the
212
-------
1500
1000.
500
220'
200
ISO.
TKN
20° C
39°0
NH4-N
20°C
3S°0
05 10 15 20 25 0 5 10 15 20 25
TIME hours
200
N03-N
20° C
0 5 10 15 20 25 05
TIME hours
10 15 20 25
FIGURE 71
DENITRIFICATION AT 20°C AND 35°C
RUN II
213
-------
140
100-
fc
o>
2501
I
200-
N03-N
NOg-N
ZO'C
0 20
S. -150
100-
50-
15
10
TIME - hrs., BOD, COD
0 5 10 15 20 25
NH4-N
300-
250-
o>
= 200-
01
ZO'C
150
100
50
20 40 60 60
TIME - hours
TKN
2o'c
COD
BOD
35'
20*C
4000
3000
.2000 i
§
O
•1000 g"
CD
0 20 40 60 80
TIME - hrs. TKN
FIGURE 72
DENITRIFICATION AT 20°C AND 35°C
WITH MANURE AS AN ADDED HYDROGEN DONOR
RUN III
214
-------
retardation of denitrification that became apparent during the longer run
of the experiment.
d) Rim. IV. - Effective denitrification is known to occur in activated
sludge at the expense of endogenous hydrogen donors. In the previous
runs complete denitrification did not occur. It was possible that the
endogenous reserves were not present in sufficient amount in the micro-
bial mass contained in the system. To investigate whether higher rates
and degrees of denitrification can be achieved by increasing the mixed
liquor suspended solids, the following experiment was conducted.
From an actively nitrifying unit, 3.6A of mixed liquor was centrifuged
and 2.4X, of the supernatant was carefully withdrawn to obtain a three-
fold concentration of MLSS. The MLSS and the total solids in the resus-
pended concentrated mixed liquor were 3430 and 4775 mg/1, respectively.
The degree of denitrification increased both at 20 and 35°C (Figure 73).
The higher quantity of endogenous material made available by concen-
trating the mixed liquor had an effect on denitrification. The plateau,
observed in the removal of nitrate in the previous run, was observed in
this run, not only at 20°C, but also at 35°C although it existed for a
shorter period in the latter instance. All the oxidized nitrogen could
not be removed from the system even at the higher MLSS level. Only 45
and 70% of the oxidized nitrogen was removed at 20° and 35°C, respectively,
in three days. The overall rate of oxidized nitrogen removal in this
run was 0.3 and 0.4 mg/1 N/g of TS/hour and 0.4 and 0.6 mg N/g of MLSS/
hour at 20°C and 35°C.
Higher denitrification rates were observed in the earlier stages of the
run. The higher MLSS did result in a significant increase of NH.-N in
the units.
e) Rur^V_- In the continuous flow nitrification units it was observed
that the formation of nitrite was significantly higher than nitrates
under high TKN loadings and that the nitrites persisted without further
oxidation. It was decided to conduct an experiment to study the pro-
gression of denitrification in the highly nitrified mixed liquor using
the mixed liquor oxygen demand as the hydrogen donors. The previous
runs had used mixed liquors high in nitrates. Two denitrification
reactors with unit E1 mixed liquor containing high nitrites were set up,
one at 20°C and the other at 35°C. The MLSS concentration was 5300
mg/1 and the results obtained were presented in Figure 74.
Over a six day period, the N02-N loss in the 20°C reactor was 0.4 mg
N/g SS/hour. The N03-N loss was 0.18 mg/g of MLSS/hour over a 24 hour
period. After the 24 hour period no NOg-N was lost from the 20°C system.
The denitrification in the 35°C reactor occurred at a much faster rate
and the rates of N02-N and N03-N computed over a three day period were
0.82 mg N/g SS/hour and 0.16 mg N/g SS/hour. In this run, no significant
change in the TKN occurred.
215
-------
eso
N03-N
20 40 60 80
2.6
2.0
1.6-1
1-0-1
N02-N
TIME
3501
300-f
250-1
200-1
ISO
o>
£ 100 J
50-
TKN
35" o
ZO'C'
3000-1
2500
COD
10 20 30 40 50 0 10 20 30 40 50
TIME - hours
FIGURE 73
DENITRIFICATION AT 20°C AND 35°C
WITH ENDOGENOUS HYDROGEN DONORS
FROM A NITRATIFIED MIXED LIQUOR
RUN IV
216
-------
350
300
250-
20O-
150
o>
IOO-
50
N02-N
1401
120-
100
N03-N
400
300
200-
100
TKN
NH4-N
246
TIME - days
FIGURE 74
DENITRIFICATIQN AT 20°C AND 35°C
WITH ENDOGENOUS HYDROGEN DONORS
FROM A NITRITIFIED MIXED LIQUOR
RUN V
-------
The total nitrogen remaining at the end of the six day denitrification
period was 56% and 50% in the 20°C and 35°C reactors respectively. Most
of the nitrogen remaining in the units was in the form of unnitrified TKN,
The plateau observed in the previous denitrification runs occurred in
this run also, suggesting that it might be a constantly occurring phe-
nomenon in the denitrification of nitrified poultry manure wastes.
Studies on the denitrification of municipal sewage effluents and agri-
cultural wastewaters showed a close correlation between predicted and
observed removals of COD and oxidized nitrogen (63). In the studies,
the following stoichiometric relationship was developed to predict the
requirement of methanol which was used as a hydrogen donor.
Cm = 2.47 N03-N + 1.53 N02-N + 0.87 DO (81)
All units in the above equation are in mg/1 of the respective components
and C is the methanol requirement.
The above equation can be used with other hydrogen donors by converting
the methanol requirement into oxygen equivalents. Since one pound of
methanol has a theoretical oxygen demand of 1.5 pounds, the above equation
can be expressed in terms of oxygen demand as
OD = 3.7 N03-N + 2.3 N02-N + 1.3 DO (82)
where all the items are in mg/1. From this equation it can be calculated
that 1000 mg of COD/1 is needed to denitrify 270 mg/1 of N03-N or 435
mg N02-N/1 assuming that no dissolved oxygen is present in the system.
Using these equivalents the predicted COD decrease corresponding to the
observed decrease in N03~N and N02-N of this run was computed and com-
pared with the actual COD decrease in the denitrification reactor. These
computations with the COD, N03-N and N02~N are presented in the Appendix,
Table VII. Figure 75 represents the theoretical and observed COD
decreases based on the N02-N plus N03-N removed from the system. The
theoretical COD decrease for the observed N02-N decrease was compared
with the observed COD decrease in the same Figure.
The removal of N02-N and N03-N together account for the decrease of COD
reasonably well, rather than to the N02-N alone. Thus in the removal of
N02-N and N03-N, the oxygen demand exerted by the endogenous hydrogen
donors contained in the mixed liquors was related to the oxidized nitrogen
as predicted by Equation 82.
218
-------
INi
IO
Ld
t/5
<
UJ
CE
(_>
UJ
Q
Q
O
O
UJ
1000
800-
600-
400-
200-
3-35'C, OBSERVED
35°C, THEORETICAL
e
20°C, THEORETICAL 80°'
o
20°C, OBSERVED 600-
400-
200-
days
FIGURE 75
COD DECREASE RELATED TO OXIDIZED
NITROGEN DECREASE
RUN V
35°C, OBSERVED
35°C, THEORETICAL
e>
£0°C, THEORETICAL
o
20°C, OBSERVED
-------
f) Run_VI_- In the previous denitrification runs, the phasic removal of
nitrite and nitrate was constantly observed. To explain this, it was
hypothesized that the initial rapid removal of the N02-N and N03-N was
perhaps due to the availability of readily assimilable hydrogen donors.
When these were exhausted the rate of denitrification decreased causing
a plateau. During the period of the plateau, a microbial population was
presumably adapting to the complex hydrogen donors present in the system.
When once a population developed that could utilize these complex donors,
denitrification again proceeded. To test this hypothesis the following
experiment was set up involving three denitrification reactors.
Reactor I - The mixed liquor from a nitrifying unit was centrifuged and
the supernatant discarded. The settled solids were harvested and washed
twice in 0.005M phosphate buffer to remove the adhering supernatant and
suspended in a mineral salts medium. Sodium nitrate, sodium nitrite,
and ammonium chloride were added to bring up the level of nitrite, nitrate,
and ammonia nitrogen to approximately the original concentrations.
Glucose was added as the readily assimilable hydrogen donor to the recon-
stituted mixed liquor to result in a concentration of 1000 mg/1. In the
following graphs, data from this reactor are noted as glucose plus cells.
Reactor II - To another aliquot of the nitrifying mixed liquor, glucose
was added to result in a concentration of 1000 mg/1 of glucose as a sup-
plemental hydrogen donor in addition to the hydrogen donors already present
in the poultry manure mixed liquor. This reactor was used to find out
whether glucose addition has any additive effect in terms of bringing
about faster denitrification. The data from this reactor are noted as
glucose plus chicken manure (CM).
Reactor III - This reactor contained the nitrifying mixed liquor, was
used as a control, and the data noted as chicken manure (CM).
Besides the routine general analyses in this run, the COD of both com-
posite samples and the filtrate, obtained by filtering the samples
through 0.45y filter paper, was determined. The COD of the filtrate
was determined primarily to find out whether the soluble portion of the
COD is preferentially used during the denitrification. The results are
presented in Figures 76 and 77.
No phasic removal of N02-N and N03-N was observed in the reconstituted
mixed liquor containing glucose ( Reactor I). However, such a phasic
removal occurred in the mixed liquor supplemented with glucose (Reactor II)
as well as in the control. The addition of a readily available substrate
such as glucose to the mixed liquor as such did not eliminate the plateau
but did increase the rate of denitrification. It is very likely that the
nature of the supernatant and not the composition of the microbial mass
itself dictates the occurrence of such a plateau.
The COD data on the filtrate indicated that the soluble COD of the glu-
cose treated systems, (Reactors I and II) decreased rapidly with a
220
-------
300'
-
200-
.
o
itm
^~
j_
o
Q.
O>
E
100-
0.
N02-N
N03-N
o GLUCOSE -t- CELLS
* e GLUCOSE + CM
I A CM
P^-- A
| A
! \
i \
i \
^ — x
X
X
1 X
i s
1 x
1 X
1 X
1 X
)' X
\ X
1 \ s
e\ ^
'' X>,
i \
i \
i \v
\ X
x x
cp \ Xx
/ "\^- -__Xxx
aZ».^»™_fi«JV^—---™ X<^
r^SPPHi^fj1 11 A^ ^ ~*^ i ( u^fKUtZ ( ( ( f
024 6 8 1C
TIME - days
FIGURE 76
DENITRIFICATION DUE TO
ENDOGENOUS AND EXOGENOUS HYDROGEN DONORS
RUN VI
221
-------
3000
2000-
CJ>
O
O
O
1000
COMPOSITE
FILTRATE
o GLUCOSE 4- CM
« GLUCOSE 4- CELLS
CM
0
246
TIME - days
FIGURE 77
COD REDUCTION IN DENITRIFICATION
DUE TO VARIOUS HYDROGEN DONORS
RUN VI
222
-------
concomitant decrease in N03~N. There was no significant decrease of the
COD in the filtrate of the control indicating that whatever decrease in
the N03-N observed in Reactor III was primarily due to the utilization
of COD from the suspended solids fraction of the mixed liquor.
The increase in nitrite in Reactor I (Figure 76) undoubtedly was due to
the rapid reduction of nitrate during denitrification. After most of the
readily assimilable hydrogen donors were utilized in Reactor II, its
rate of denitrification was comparable to the control (Figure 76). The
results of this study indicate that the occurrence of the plateau in the
denitrification experiments with poultry wastes are likely due to
the nature of the supernatant of the mixed liquor. Even though these
experiments were conducted in laboratory units, similar plateaus were
obtained in exploratory pilot scale systems. Therefore, such plateaus
are likely to occur in field units denitrifying poultry manure waste-
waters. The addition of a readily assimilable hydrogen donor increases
the rate of denitrification minimizing the effect of the mixed liquor
supernatant on the plateau and increasing denitrification in the inital
period.
g) Run VII - In Run V, it was noted that the stoichiometric relation-
ships between the COD decrease and the N02 + N03-N decrease were valid
in a highly nitritifying system. Run VI noted that oxidized nitrogen
can be removed faster by supplementing with a readily available substrate
such as glucose. In this experiment we have considered both the above
aspects and attempted to denitrify the highly nitratifying mixed liquor
of unit F1 by supplementing it with glucose and poultry manure separately
at two levels over and above the oxygen demanding materials present in
the mixed liquor.
The supplemental amounts of glucose and chicken manure were determined
by assuming their theoretical oxygen demand and calculating the required
amount to remove the amount of N03-N present in the mixed liquor based
on the stoichiometric relationship indicated previously (Equation 82).
The stoichiometric amounts of these substrates and twice these amounts
were added to respective reactors. A control without any addition of
these substrates was included. The reactors with added chicken manure
had considerably higher solids than did the other reactors. The mixed
liquor contained negligible amounts of nitrite.
The computations for glucose and poultry manure are presented in Appendix,
Table VIII. The removal of N03-N, total nitrogen, and COD are presented
in Figures 78 and 79. The rates of denitrification are presented in
Table 33.
The denitrification rates were low although hydrogen donors were added
in addition to the ones already present in the system. The rate of
N03-N loss in the control was lower than the rate observed in the control
of Run V (0.067 vs 0.18 mg N03-N/g SS/hour).
223
-------
ro
ro
100
CONTROL
CONTROL a
GLUCOSE (IX)
CONTROL 8
GLUCOSE (2X)
CONTROL a
CHICKEN MANURE (IX)
CONTROL a
CHICKEN MANURE (2X) 150 -
TIME
FIGURE 78
NITROGFN PATTERNS IN DENITRIFICATION
DUE TO EXCESSIVE HYDROGEN DONORS
RUN VII
2501
200-
150-
100-
50-
0
\'
x
K
\
ILW • -*Sr-^
\^^^— — o
\ 0
^ .
1 1
0
cc
i-
o
o
LU
I
1-
U_
O
^™ III
ll
LU
LU
2L
r-
h- 0
Z"LU
0 ^
EH
^LJ ^^
CL ^
0 5 10 15
days
-------
16000 n
_ 12000-
o>
8000-
o CONTROL ft CHICKEN MANURE (2X)
• CONTROL 8 GLUCOSE (2X)
a CONTROL 8 GLUCOSE (IX)
• CONTROL
0
O
U
4000-
I I r
4 8 12
TIME - days
16
FIGURE 79
COD REMOVAL PATTERNS WITH
DIFFERENT EXOGENOUS SUBSTRATES -
DENITRIFICATION UNIT
RUN VII
225
-------
rv>
Reactor
TABLE 33
EFFECT OF THE ADDITION OF EXOGENOUS
HYDROGEN DONORS ON DENITRIFICATION
RATES OF POULTRY WASTE
Solids Content n .•i..j£j«,a.J__ r
Control (mixed
liquor from
Unit F1 )
Control +
glucose (lx)
Control +
glucose (2x)
Control + 1
chicken manure(lx)
TS
8290
8310
8740
2450
Control + 16040
(mg/1)
SS
4790
4750
4830
8330
10740
UCII 1 1,1 1 1 1 l*C
N03-N/g TS/hr
0.039
0.053
0.18
0.1
0.1
I U I UN l\Ct WC
N03-N/g SS/hr
0.067
0.093
0.32
0.15
0.16
Time considered in
computing rate (days)
7
13
7
7
7
chicken manure(2x)
-------
The NCL-N concentration in the reactors was 322 mg/1. From the COD
removal (Figure 79) and the N02-N removal (Figure 78) data, it can be
seen that the stoichiometric relationship, i.e., the requirement of
1000 mg/1 of COD removal for each 270 mg/1 of N03-N removed, does not
seem to hold in this case. More COD was removed than predicted by N03-N
removal alone. Other COD removal mechanisms were operative. This is in
contrast to the relationship observed in the highly nitritified system
(Run V).
It appears that the nature of the highly nitratified mixed liquor in
Unit F1 had an inhibitory effect on the rate of N03-N removal in these
experiments. Even after thirteen days of denitrification, about 50% of
NO.,-N still remained in the reactor with Ix glucose theoretically needed
to bring about complete denitrification. Nearly complete removal of N03-N
was accomplished in other reactors where adequate oxygen demand occurred.
However, the units treated with chicken manure contained considerably
higher amounts of total nitrogen than the control (Figure 78) as would
be expected. In the units where the chicken manure was added, the remain-
ing nitrogen was due to the TKN added.
The results of this study indicate lower rates of denitrification with
highly nitratified mixed liquors (control). Addition of hydrogen donors
such as glucose and poultry manure increased the denitrification rates.
The observed decrease in COD was more than that predicted by N03-N
removals indicating that besides N03-N removal, other COD removal mecha-
nisms were operative. It is undesirable to add untreated poultry manure
as a hydrogen donor for denitrification because the total nitrogen of the
system is increased rather than decreased.
h) Run VIII - In Run VII the effect of added hydrogen (poultry manure
and glucose) on the denitrification of a highly nitritifying mixed liquor
was studied. In this experiment, the effect of supplemented hydrogen
donors on the denitrification on the highly nitritifying mixed liquor of
unit G1 was evaluated. The N02-N concentration was 885 mg/1 and the
N03-N concentration was negligible.
The levels of glucose and chicken manure needed theoretically to achieve
complete denitrification were calculated as before. The amounts of
glucose added were approximately one and two times the calculated stoichio-
metric amount required. In the reactors receiving the supplemental poultry
manure, the actual levels were 0.75 and 1.5 times the calculated theo-
retical requirement (Appendix Table VIII).
The results of this run are presented in Figures 80 and 81. Although a
readily assimilable hydrogen donor, such as glucose, was added at an
adequate level, rapid initial denitrification could not be accomplished.
227
-------
ro
ro
oo
100
CONTROL
CONTROL a 2g/l
GLUCOSE
CONTROL 8 4 g/l
GLUCOSE
CONTROL a 32.86 g/l
CHICKEN MANURE
CONTROL a 65.79 g/l
CHICKEN MANURE
250
200
O
O
UJ
-100
M
Q
UJ
•=50
o
days
10
15
20
FIGURE 80
DENITRIFICATION OF UNIT G1
MIXED LIQUOR WITH EXOGENOUS
HYDROGEN DONORS
RUN VIII
-------
10 19
TIME
0
doys
CONTROL a -4g/l
OLUCOSE
OBSERVED
I 4000
THEORETICAL
£ fiCOO
CONTROL a 3Z.B69/I
CHICKEN MANURE
10 19
TIME
CONTROL a 657g/l
CHICKEN MANURE
0 9
days
FIGURE 81
THEORETICAL AND OBSERVED COD
DECREASES - DENITRIFICATION RUN VIII
229
-------
A similar observation was made when chicken manure was used as a hydrogen
donor. This suggests that there might be inhibitory factors suppressing
the activity of the facultative denitrifiers present in the nitrifying
mixed liquor and that rapid denitrification takes place when the popula-
tion was adapted to the environmental conditions. It is possible, but
less likely, that the population of the denitrifiers was small initially
and that it took a long time to establish an adequate population to bring
about denitrification.
There was an apparent increase in the percent TN remaining of the control
(Figure 80) in the case of the reactor containing 65.8 g of chicken manure
per liter. This was due to the inhibition of denitrification caused by
excessive chicken manure added. In the control, denitrification occurred
whereas in the control plus 65.8 g chicken manure/liter reactor, denitri-
fication did not occur to the same extent. Hence the computation of per-
centage TN remaining of the control for this reactor showed an increase.
The observed COD decrease and the theoretical decrease of COD due to the
denitrification of oxidized nitrogen were computed as in Runs V and VII
and were plotted in Figure 81. There was good agreement between the
observed and theoretical COD:N decrease in the control reactor, an obser-
vation similar to the one made in Run V using a highly nitritifying mixed
liquor. However, there was poor agreement between these relationships
in the reactors treated with either glucose or chicken manure. In these
reactors the observed COD decrease was considerably higher than the
theoretical value. Only a fraction of the COD decrease was used by the
denitrifying microorganisms as a hydrogen donor. It is likely that most
of the COD decrease observed in the control was due to the metabolism of
the readily available substrate in the mixed liquor as well as the
exertion of the endogenous oxygen demand.
i) Rim D(_ - In the previous run it was inferred that addition of high
quantities of chicken manure (65.8 g/liter) inhibited the denitrifica-
tion of the nitrified mixed liquor. To find out at what level of dosage
inhibition may occur, a batch experiment was set up in which the highly
nitritified mixed liquor of unit G1 was used. The N02-N concentration
was 880 mg/1 and the NOg-N concentration was negligible. Poultry manure
was added at 35.7, 42.9, 50, and 57.2 g/1 levels to aliquots of unit 6'
mixed liquor and blended. After portions of these homogeneous suspen-
sions were saved for initial analysis, the remainders were denitrified.
A compendium of the results obtained in this run and Run VIII are pre-
sented in Figure 82. For comparative purposes the results are presented
on the basis of the ratio of the amount of wet manure added to the amount
of oxidized nitrogen in the units. In Run VIII, the inhibition occurred
at a ratio of 77.3:1 but not at a ratio of 37.4:1. This experiment con-
tained reactors having ratios spanning this range.
Denitrification was not inhibited at a wet TS:N02-N ratio of 72.1,
(Run IX) but it was inhibited at 77.3:1 (Run VIII) level. The raw
poultry manure added in these two runs were obtained on two different
days and differed significantly in their relative solids contribution.
230
-------
1000
A —
A -
o —
37.4
42.7
52.4
60.5
72.1
77.3
25
15 20
TIME - days
FIGURE 82
DENITRIFICATION AT VARIOUS
WET MANURE LOADINGS -
DENITRIFICATION RUN IX
The solids content of the poultry manure used in Run VIII was less than
that used in Run IX with the result that the reactors in these runs
contained significantly different dry solids concentrations (Table 34).
The higher amount of solids added on a dry basis in Run IX did not
inhibit the denitrification. Thus on the basis of the weight of dry
solids, there was no correlation between solids and any inhibition of
denitrification. It is likely that the inhibition observed in Run VIII
may have been due to factors present in the wet manure. These factors
may have been lost under the storage and drying conditions that produced
the drier manure. The inhibition also could have been due to other
intrinsic differences between the two batches of manure used in Run
VIII and IX. Because of the high concentrations necessary to possibly
cause inhibition of denitrification, such inhibition is inconsequential
to the practical application of denitrification of poultry manure sus-
pensions. Reasonable rates of denitrification can be accomplished by
smaller concentrations of untreated poultry waste and other exogenous
hydrogen donors.
231
-------
TABLE 34. DRY SOLIDS CONTENT OF WET MANURE
USED IN DENITRIFICATION STUDY
Reactor*
Weight of wet
manure added
(gm/1)
Added dry
solids
(gm/1)
Ratio of wet solids to
NO,
N03-N
D1
E1
B
C
D
E
32.9 g/1
65.7 g/1
35.7 g/1
42.9 g/1
50.0 g/1
57.2 g/1
4.69
9.95
16.06
19.82
25.87
28.35
37.4:1
77.3:1
42.7:1
52.4:1
60.5:1
72.1:1
Reactors D1 and E1 were used in Run VIII; the other reactors were
used in Run IX.
j) Run X - In determining the effect of dilution on nitrification and the
oxygen uptake of a chicken manure mixed liquor, (1.91% total solids), 1:1
and 1:2 diluted suspensions exerted progressively higher oxygen uptake
than the undiluted mixed liquor (Figure 61). It was inferred from these
studies that the observed increase in oxygen uptake was due to allevia-
tion of toxicity and increased nitrogenous oxygen demand. To find out
whether the dilution of the nitrified mixed liquor will yield higher
rates of denitrification because of increased oxygen demand and alle-
viation of toxicity, a denitrification experiment was set up. An undi-
luted oxidation ditch mixed liquor (1.91% total solids) and 1:2 diluted
suspension of the same were used. This oxidation ditch mixed liquor
contained relatively high N02-N and low N03-N. The effect of N02-N,
N03-N, and NH4~N on the denitrification of the diluted suspension was
also studied by supplementing the diluted system with sodium nitrate,
sodium nitrite, and NH.C1. The results are presented in Figures 83
and 84. H
A similar denitrification experiment was performed with the oxidation
ditch mixed liquor when its total solids concentration was 6.3%. The
N02-N concentration was approximately 1200 mg/1 and there was no sig-
nificant amount of N03~N. In this experiment, a 1:3 diluted mixed liquor
was also denitrified to find the effect of dilution on the rate of deni-
trification. The results are presented in Figure 85. The rates of
denitrification observed in these two sets of experiments are given in
Table 35.
There was no significant change in the denitrification rate because of
dilution either in the case of mixed liquor containing 1.91% total solids
232
-------
11000
2500
FIGURE 83
DENITRIFICATION OF DILUTED
NITRITIFYING MIXED LIQUOR
RUN X
233
-------
5000-
E 4OOO-
o
o
o
3OOO-
2000-
1000
l'2 ML » N02-N * N03-N
400
CT>
300-
200-
100 -
N02-N
0 1=2 ML + NOg-N+ N03-N
4
6 12
TIME
1=2 ML+N02-N +N03-N + NH4-N
FIGURE 84
DENITRIFICATION OF DILUTED NITRITIFYING
MIXED LIQUOR SUPPLEMENTED WITH NITROGEN
RUN X
234
-------
CM
1200
1000
800-
600-
400-
200-
'AS IS' ML CONCENTRATION
TOTAL SOLIDS '• 63000 MG/L
SUSPENDED SOLIDS • 51400 MG/L
'/4 OF 'AS 6' ML CONCENTRATION
70,0001
Q
8
30,000-
K)pOO-
ML '• 'AS IS'
• TN (MEAN TKN + N02-N)
• MEAN TKN
o COD
ML ' '/4 CONCENTRATION
• TN
• TKN
a COD
I 2
TIME
3 4
days
[3500
•2500
pISOO
. 500
r
•
FIGURE 85
DENITRIFICATION OF A CONCENTRATED
NITRITIFYING MIXED LIQUOR
RUN X
235
-------
ro
OJ
TABLE 35. EFFECT OF DILUTION AND SUPPLEMENTED N02~N, NOg-N,
AND NH4-N ON THE RATE OF DENITRIFICATION - RUN X
SJllds Ssol?2sed .Denitriti- Denitrati- Denitrifi- Time considerec
fication rate TI cation rate cation rate • _•_•
in umt<; in iim1"<; i».«««.i»»ii •»v<_ ,^^^-.^,, • »*..>_ ,,— -.«. .,.„>. rnmniiTina
in u 1 1 1 1*0 ill uniuo Mf\ W nav* MO M nay MO -tMO M nav* v*v/iiijjuuiny
A.
B.
C.
D.
E.
F.
g TS/ g SS/ g TS/ g SS/ g TS/ g SS/
hr hr hr hr hr hr
"as is" ODML I 19,100 15,650 0.1 0.11 0.05 0.06 0.15 0.17
1:2 ODML I 6,370 5,220 0.09 0.11 0.05 0.06 0.14 0.17
1:2 ODML I + 7,940 5,220 0.13 0.2 0.04 0.06 0.17 0.26
N02-N + N03-N
1:20DMLI+ 8,260 5,220 0.13 0.2 0.03 0.04 0.16 0.24
N02-N + N03-N +
NH4-N
"as is" ODML II 63,000 51,400 0.25 0.3 -- — 0.25 0.3
1:3 ODML II 15,750 12,850 0.25 0.3 ~ — 0.25 0.3
12
12
14
14
3
3
-------
or in the mixed liquor containing 6.3% total solids. If the observed
increases of oxygen uptake of the diluted suspensions (Figure 61) were
due to carbonaceous demand, higher denitrification rates should have
resulted with the dilute mixed liquor. However, such higher rates were
not obtained in the current study confirming the conclusions drawn
elsewhere that the additional oxygen uptake exerted by the diluted
mixed liquor was not due to carbonaceous but to nitrogenous demand.
The rates of denitritification increased about 50% by supplementing the
mixed liquor with NC^-N and NCL-N. However, there was no change in the
denitratification rate. Addition of NH»-N at the concentration used in
this study did not affect the denitritification rate, but the rate of
denitratification appeared to have been reduced. In these denitrifica-
tion experiments, the phasic removal of nitrite and nitrate again was
observed.
It was concluded that higher mixed liquor concentration's increased the
rates of denitrification. It appears that the rate of denitritification
was increased by supplementing the mixed liquor with NCL-N and N03-N.
No such increases were found with denitratification rates.
k) Run XI - The effect of pH on the denitrification rates was studied in
this batch experiment. Oxidation ditch mixed liquor high in solids
(63,280 mg/1) as well as nitrite concentration (1700 mg/1) was used.
The denitrification protocol was the same except helium was used for
sparging to remove dissolved oxygen at the beginning of the experiment.
Four aliquots of the mixed liquor were adjusted to pH 4, 8, 10 and 11
rapidly to ensure minimum loss of ammonia at higher pH values and were
transferred to denitrification reactors. Another denitrification reactor
containing the untreated mixed liquor (pH 6) was kept as a control. The
reactors were not buffered intentionally to observe the change in the
initial pH. Routine analyses of the van'ous forms of nitrogen were per-
formed. A qualitative analysis of the head gases collected in the denitrv
fication reactors was performed by gas chromatography. A dual channel
Varian 200 gas chromatograph containing a stainless steel column, 12'xl/8"
packed with Poropak (100-120 mesh) and equipped with flame ionization
and micro-cross-section detectors was used. The detector temperature
was 150°C and the injector and column were operated at room temperature.
Head gases from the denitrification reactors were taken by syringes.
After the gas samples were taken for gas chromatography, the reactors
were sparged with helium to expel all the gases produced previously.
Thus the data presented in Table 36 represents the relative proportions
of gases produced between consecutive sampling dates.
The nitrogen data is presented in Figures 86 and 87. The computed rates
of denitrification are given in Table 37.
237
-------
TABLE 36. RELATIVE PROPORTION OF SOME GASES CONTAINED IN THE
HEAD GASES OF THE DENITRIFICATION REACTOR -
DENITRIFICATION RUN XI (ml/100 ml of head gas)
Initial pH 4
Day of
Experiment
1
3
5
7
9
12
16
1
3
5
7
9
12
16
_
1.7
46.6
46.6
9.2
19.8
43.6
8.9
6.6
8.4
8.0
4.3
5.4
1.8
6
3.2
17.1
7.0
25.2
16.1
10.4
41.0
6.3
7.5
7.0
8.2
11.3
18.4
8.9
8
N2
43.8
12.2
26.2
22.8
6.4
6.4
32.4
co2
8.6
6.6
6.9
12.3
10.0
10.6
8.1
10
50.4
33.8
13.3
9.2
52.0
6.6
19.4
1.0
4.2
5.2
2.2
0.4
10.0
8.4
1
46
43
28
27
19
15
15
0
0
0
1
0
20
6
1
.8
.3
.6
.2
.4
.0
.0
.6
.2
.8
.0
.6
4
8.5
3.7
5.9
6.0
1.4
2.5
0.4
0
0
0
0
0
0
0
6
17.0
11.0
15.6
9.8
0
0
0
0
0
0
0
27.2
23.0
0
8
N20
11.3
6.8
0
0
0
0
0
H2
0
0
0
21
15.2
0
0
10
6.2
7.4
12.3
16.7
0
0
0
0
0
0
0
14.6
16.8
0
11
5.0
8.0
7.4
16.4
0
0
0
0
0
0
0
0
22
48
238
-------
PH
2000
16004
500
250
to
O
FIGURE 86
DENITRIFICATION OF pH CONTROLLED
MIXED LIQUORS
RUN XI
239
-------
IN3
-p>
'O
100
o
^ 80
<
LJ
o:
60-
40 -
LU
O
£ 204
0
pH
4
6
8
10
II
8
12 16
TIME
days
FIGURE 87
NITROGEN REMAINING IN REACTORS
WITH INITIAL pH ADJUSTMENT -
DENITRIFICATION RUN XI
-------
TABLE 37. RATES OF DENITRIFICATION OBSERVED IN REACTORS*
ADJUSTED INITIALLY TO VARIOUS pH VALUES -
DENITRIFICATION RUN XI
Initial
PH
4
6
8
10
11
Denitn'tifi cation
Rate
N02-N per
g TS/
hr
0.04
0.22
0.28
0.14
0.09
g ss/
hr
0.043
0.25
0.31
0.16
0.1
Denitratifi cation
Rate
N03-N per
g TS/
hr
0.003
0.053
0.066
0.027
0.011
g ss/
hr
0.004
0.059
0.073
0.03
0.012
Denitrification
Rate
N02 + N03-N per
g TS/
hr
0.043
0.27
0.35
0.17
0.1
g ss/
hr
0.047
0.31
0.38
0.19
0.11
Days used for
computing rates
NO- N03
removal removal
14
5
4
8
14
14
5
4
10
16
* the total solids and suspended solids were the same in each flask and were 63,180 mg/1 and 56,900
mg/1 respectively
-------
The pH of the various systems changed considerably during the experi-
mental period (Figure 88). The reactors which were started at pH 4, 6,
and 8 increased in pH although those started at pH 6 and 8 first increased
and then decreased in pH. The reactors adjusted to pH 10 and 11 showed
a decreasing trend eventually reaching a pH 8-9 level. The initial
denitrification rates were significantly higher in the pH 6 and 8 reactors
than the reactors adjusted to the very low and high pH values. Never-
theless, the reactors adjusted to pH 10 and 11 denitrified after a
period of adjustment. The denitrification of the oxidized nitrogen did
not take place completely at pH 4 indicating that highly acid conditions
are detrimental. The overall denitritification rates were considerably
higher than the denitratification rates and it appears that nitrate was
removed much faster once the nitrite was removed. A similar observation
can be made from data in previous experiments (Runs VIII-X).
121
10-
6-1
-a—a—o-
a a a—a.
024 6 8 10 12 14 16
DAYS
FIGURE 88
pH CHANGES DURING DENITRIFICATION
OF pH ADJUSTED MIXED LIQUORS
RUN XI
242
-------
The total denitrification rates (Table 37) for pH 10 and 11 reactors
were computed by adding the denitritification and denitratification rates,
as was done with the rates from the rest of the reactors although the
number of days involved in the two processes were different. The dif-
ference between these computed rates and any actual rates, for practical
purposes, will be insignificant. In lieu of a better parameter for the
expression of active biomass, we chose mixed liquor suspended solids and
total solids.
Although it is shown to be feasible to remove all of the oxidized nitro-
gen from the nitrified mixed liquor, under proper conditions it can be
seen from Figure 87 that about 60 percent of the total nitrogen still
remained in the system.
The results from gas chromatography are very qualitative in nature. The
results presented in Table 36 are the relative proportions of the gases
produced with respect to the total volume of head gas injected expressed
in ml/100 ml of the head gas. The head gas, besides containing the gases
reported here, also contained water vapor, helium - the gas used for
sparging, possibly CH. and perhaps some of other gases such as NH,,
not detected by the gas chromatograph. Nitric oxide, NO, also was found
in reactors adjusted to pH 6, 8, 10, and 11, but its volume could not be
estimated because of the unavailability of a standard. Thus the volumes
of all the gases for a given reactor on a given day will not add up to
100. Illustrative chromatographic patterns are shown in Figure 89.
The data confirm that the primary end products of denitrification are
O, N2, NO, and C02, and that indeed the oxidized nitrogen forms were
removed as gaseous end products. Nitrogen gas seemed to be the major
end product. Large quantities of this gas were produced initially.
Nitrous oxide, N20, was also produced initially in significant quantities
but ceased to be formed after about a week in the reactors with an
initial pH of 6, 8, 10 and 11. However, it was produced continuously
in the pH 4 reactor. Carbon dioxide production was relatively high in
the pH 6 and 8 reactors indicating strong biological activity which
resulted in higher denitrification rates. It is interesting to note
that hydrogen was formed approximately after a week in all the reactors
except in the one initially adjusted to pH 4.
The flame ionization detector plot on the gas chromatograms suggested
the presence of a gas having a similar retention time as methane but
we did not have enough information to speculate on the identity of this
gas. This gas species occurred in the pH range of 7 to 9.
The results of this study confirmed that the gaseous end products of
denitrification were N20, N?, and NO. The optimum pH of denitrification
for these mixed liquors appeared in the pH range of 6-8. The resultant
pH of actively nitrifying poultry waste generally is in the range of
5 to 6.5.
243
-------
N20-
N
2 I-OI
1.08 N | in
NO-!1'
or
L.-.T
UJ
CO
7^
o
CL
(/>
UJ
QC
CC
UJ
Q
tr
O
o:
TIME
FIGURE 89
TYPICAL CHROMATOGRAM OF
DENITRIFICATION HEAD GASES
RUN XI
244
-------
Nitrification of the Denitrified Manure
From the batch and continuous flow nitrification experiments, it was
found that only 50 to 60% of the total nitrogen in the feed could be
nitrified, and only this amount can be removed by denitrification leaving
about 40 to 50% of the total nitrogen of the feed still in the unoxidized
form in the system. To find out whether the nitrogen left after denitri-
fication can be nitrified further, batch reactors were set up under the
following conditions. A mixed liquor which was denitrified for a period
of 23 days was used in this second nitrification step.
a) Denitrified mixed liquor plus aeration only - In this reactor no con-
ditions were changed except the previously denitrified mixed liquor was
aerated at an adequate level to produce high dissolved oxygen values.
b) Denitrified mixed liquor plus poultry manure suspension plus aeration -
680 ml of fresh poultry manure suspension (553 mg/1 COD) were added to
680 ml of denitrified mixed liquor (1092 mg/1 S.S., 1567 mg/1 COD) and
aerated. The poultry manure was added to provide nutrients and additional
nitrogen.
c) Denitrified mixed liquor plus actively nitrifying seed plus aeration -
To 680 ml of the denitrified mixed liquor, 680 ml of an actively nitrifying
mixed liquor was added and aerated. The seed was provided in the event
no nitrifying organisms remained after denitrification.
d) Denitrified mixed liquor plus actively nitrifying seed plus poultry
manure plus aeration - To 680 ml of denitrified mixed liquor 136 ml
(10% seed by volume) of an actively nitrifying mixed liquor and 680 ml
of the poultry manure suspension used in reactor b was added and aerated.
This mixture would assure adequate nutrients, nitrogen, and nitrifying
organisms and could approximate start-up conditions of an actual system
that had been denitrified. The results of this study were presented in
Figures 90 and 91.
All the systems nitrified. It was expected that the denitrified mixed
liquor, because of its anaerobic nature, would not be able to support
the growth of nitrifying bacteria. Nevertheless, the existence of a
nitrifying population, which may have been dormant during the denitri-
fication phase, was evident and it was able to oxidize the nitrogen in
the system once the aeration was resumed. There were significant losses
of total nitrogen from the system, about 26-43%, in spite of nitrifica-
tion. A similar observation has been made in nitrification studies using
batch systems.
The results of this study showed that the nitrifying population survived
the entire period of anaerobiosis during denitrification of 23 days.
This was indicated by the occurrence of nitrification in the denitrified
mixed liquor when aerobic conditions were maintained. The formation of
nitrate from nitrite in this nitrification process indicates that both
the ammonia oxidizers and the nitrate formers were able to survive
theanoxic conditions of the prolonged denitrification phase.
245
-------
2001
200
160 J
120 J
80-1
40 J
o>
a.
04 8 12 16 20 . o 4 8 12 16 20
&
50
25
I03-N
/
/
/
N02-N
REACTOR
c
04 8 12 16 20
TIME - days
FIGURE 90
NITRIFICATION OF DENITRIFIED
POULTRY WASTES
RESULTS OF REACTORS b AND c
246
-------
200
160
200
I60-i
120
80-
S 40-
04 8 12 16 20
REACTOR
a
TKN'
NH3-N
6 12 16 20
40
20-
REACTOR
a
8 12 16 20
TIME - days
FIGURE 91
NITRIFICATION OF DENITRIFIED POULTRY MANURE -
NITRIFYING ORGANISMS ADDED
RESULTS OF REACTORS a AND d
247
-------
These observations offer the possibility of utilizing cyclical nitrifi-
cation-denitrification sequences for the removal of nitrogen. In prac-
tical situations with poultry wastes, it is unlikely that problems with
respect to initiation of nitrification after denitrification will occur.
The nitrifying organisms in these experiments were not found to be fas-
tidious in that they survived the anoxic conditions prevailing during
denitrification. The inherent nitrifying ability of the microorganisms
contained in the mixed liquor denitrified over a prolonged period has
not been reported earlier. The engineering possibilities of a cyclic
nitrification-denitrification system for nitrogen control in wastewaters
also has not been reported elsewhere.
All the residual ammonia present in the mixed liquor after the denitri-
fication was nitrified in the second nitrification step. It appears that
the denitrification step alleviated the effect of any factors that were
responsible for the persistence of the residual ammonia in the initial
nitrification step.
Although nitrification occurred after denitrification, significant losses
of total nitrogen resulted in this nitrification step. If such losses
occur in an actual cyclic nitrification-denitrification sequence, such
nitrogen loss is an added benefit to nitrogen control.
Chemical Denitrification
In the batch nitrification study, a significant loss of nitrogen was
observed in reactors even when the mixed liquor had a high amount of
dissolved oxygen (Figure 56). It was hypothesized that some of the
losses could be due to volatilization of ammonia in the initial stages
of operation and also due to denitrification that might be taking place
in localized anaerobic pockets, particularly within the floe. In addition
to these possible means of nitrogen losses, under suitable conditions
nitrous acid reacts with amino acids to yield molecular nitrogen (Equa-
tion 60). In a reaction quite similar to the above, ammonia may react
with nitrous acid to yield molecular nitrogen (Equation 61). To find
out whether a part of the nitrogen losses could be attributed due to
the chemical denitrification of the nitrite, a study was undertaken with
a highly nitritifying oxidation ditch mixed liquor.
To one batch reactor containing 3 liters of oxidation ditch mixed liquor,
20 ml of a saturated solution of HgCl2 was added to kill the bacterial
population. Another reactor containing 3 liters of mixed liquor was kept
as a control. Both the reactors were purged with nitrogen gas and tightly
stoppered. Samples were drawn routinely and analyzed for N02-N. Figure
92 represents the amount of N02-N remaining in the units with time. The
results indicate that the nitrogen losses due to chemical denitrification,
if any, are negligible.
248
-------
1200-
1000-
o> 800-
600-1
z
I
CVJ
Z 400J
200-
OXIDATION DITCH ML
& HgCI2
OXIDATION DITCH ML
CHEMICAL
DENITRIFICATION
EXPERIMENTS
TIME - days
FIGURE 92
EFFECT OF MERCURIC CHLORIDE
ON DENITRIFICATION
-------
SIGNIFICANCE OF THE RESEARCH
General - The results of this study indicated that it is feasible to
reduce the level of nitrogen in the poultry manure by subjecting it to
a sequence of microbial nitrification and denitrification. Some of the
important parameters that govern the processes of nitrification and
denitrification in poultry manure were studied on a laboratory scale
and the following discussion pertains to these studies.
Nitrification
Seed - Since nitrification in an ecosystem is primarily the result of
the activity of a specialized group of autotrophic organisms, it is
imperative that conditions for their growth and survival be optimal
to sustain the process. In the waste treatment processes it is not
critical that a particular species of heterotrophic organisms be present
to achieve a high degree of BOD removal since many heterotrophic organisms
have the ability to metabolize organic wastes. However, to achieve nitri-
fication, an active nitrifying population, which is comprised of only a
few species of organisms in nature, has to be maintained. Raw poultry
manure does not harbor the nitrifying flora. This study indicated that
poultry wastes can not be nitrified without the addition of a nitrifying
seed thus confirming the observations of other investigators (120).
In an in-house oxidation ditch treating poultry waste from about 250
chickens, nitrification apparently occurred although the ditch was con-
structed with concrete and started without any bacterial seed (99). In
this case, the nitrification perhaps was induced by the soil and dust
swept from the floor into the oxidation ditch.
Active nitrification can be made to take place quickly by adequately
seeding the units with a highly nitrifying mixed liquor. The quality
of the seed material can make a significant difference in the resultant
end product of nitrogen oxidation. In a unit seeded with soil (unit A)
only nitrite resulted whereas in unit B1 which operated approximately
under the same conditions but was seeded with nitrifying activated sludge,
the end product was nitrate. The soil used to seed unit A apparently did
not contain nitrite oxidizing bacteria.
In pure culture studies the size of a Nitrobacter inoculum seemed to play
an important role in the oxidation of nitrite (147). A time of 150 hours
was required for the oxidation of 140 mg/1 of nitrite nitrogen contained
in 200 ml of culture medium by an inoculum containing 7 x 107 organisms
whereas the same amount was oxidized completely in 50 hours with 5 times
the number of organisms. The multiplication of the organisms contained
in the smaller volume of the inoculum was delayed due to the toxicity of
the nitrite in the medium. The same nitrite concentration did not have
any apparent toxic effect when the population was large. Pure culture
studies have shown that very high levels of nitrites (5000-6000 mg/1) were
oxidized provided the highest level was reached in steps (147) and the
Nitrobacter population was in its log growth phase.
250
-------
Units in this study having high TKN loadings tended to accumulate nitrites.
These units were not seeded periodically and it will be of future interest
to find out whether such a seeding of the units with a seed containing a
large population of nitrobacters would result in the complete oxidation
of nitrite. It is highly unlikely that the detention time was a factor
in the incomplete oxidation of nitrite because in the units E and F the
SRT was very high, 465 and 870 days respectively, and yet nitrite con-
centration was much higher than the nitrate.
Inhibition - The nature of the oxidized nitrogen species formed can be
related to the loading of the unit. The TKN loading of the units appears
to be an important factor. From the continuous flow unit data, it appears
that nitrite was the primary end product when the TKN of the feed was
approximately 600 mg/1 or higher.
Total Kjeldahl nitrogen of the feed is not of itself likely to be a cause
of inhibition of nitrification. Rather the end products of TKN metabolism,
such as ammonia or nitrites, are a more likely cause of any resulting
inhibition if their concentrations in the media are large enough.
The accumulation of nitrite was also noticed with a sewage fed activated
sludge when the NH.-N concentration was increased from 320 to 480 mg/1.
At these NH4-N values the BOD:N ratios in the raw feed were 0.3:1 and
0.2:1 respectively (186). However, with an activated sludge fed a syn-
thetic feed, the nitrite concentration in the effluent tended to increase
when the NH^-N was higher than 112 mg/1, corresponding to a BOD:N ratio
of 3.2:1. The difference between the behavior of the two sludges was
attributed to the higher resistance of the bacterial population in the
sewage fed activated sludge to higher concentrations of NH.-N in sewage.
The higher concentration of nitrite in the effluents at higher ammonium
loadings was believed due to a deficiency of BOD rather than any specific
toxic effect of the ammonium ion. Knowledge of the nitrification process
indicates that accumulation of ammonia or nitrite may be the reason for
the inhibition of nitrification rather than a deficiency of BOD. Our
studies also indicated that BOD deficiency was not an important factor
in the inhibition of nitrification.
In pure cultures it has been reported that free ammonia (89, 105, 147,
148) and undissociated HN02 (146) were more inhibitory to nitrite oxida-
tion than the NH.+ or N02~ concentration per se. The dissociation
equations for ammonium hydroxide and HN02 indicate that the concentration
of free ammonia increases with an increase in the pH, and the concentration
of undissociated nitrous acid increases with a decrease in pH. In our
study we have attempted to find the effect of these undissociated forms
on the oxidation of ammonium and nitrite.
The concentration of free NH, and undissociated HN02 were calculated from
the following equations:
251
-------
17 mgNH/-N/l x 10pH
Free NHV mg/1 = U- x = rn (83)
Kb/Kw + 10
undissociated HN09 = if-x 10(3-4'pH) x NO--N , (84)
(mg/D 2 14 L .
Equation (83) was developed in the previous part of this report dealing
with ammonia desorption. Equation (84) was taken from reference 146.
If free ammonia and undissociated nitrous acid do inhibit nitrification,
then relationships relating the concentration of these compounds to the
amount of nitrification and/or relating the concentration per unit of
active nitrifying organisms to the amount of nitrification would be
expected to indicate critical levels of these factors.
Figure 93 shows the relationship between the free NHg/total volatile
solids and the percent nitritification and nitratification respectively
using the data from the continuous flow units without recycling. The
relationship between the free HNCL/total volatile solids and the percent
nitritification and nitratification is presented in Figure 94. These
relationships indicate that at low ratios of undissociated NH7/total
. >• - • J
volatile solids and HN02/total volatile solids, the oxidized form of
nitrogen was predominantly nitrate whereas at the higher ratios inten-
sive nitrite formation was the result. The pattern resulting from undis-
sociated HNCL (Figure 94) was less definite probably because the nitrate
formation was more inhibited by NH.-N than by HNCL.
It should be noted that these relationships were found at SRT values
long enough to keep an active nitrifying population within the system."
At an SRT of approximately 1 day, no net nitrification took place because
of the washout of the nitrifying population from the units. Under these
conditions, high concentrations of free NH~ resulted because of the
relatively high pH of the system.
At high COD or TKN loadings, the concentration of free ammonia and undis-
sociated nitrous acid was high and N02-N was the primary end product with
comparatively small amounts of N03-N formed. The high nitrite formation
was observed within a loading range of 0.13 to 1:1 based on a COD:MLVSS
ratio. Higher loadings also should produce high nitrites. Nitrate was
the primary product of nitrogen oxidation below the 0.13:1 loading factor,
Assuming that in raw chicken manure there are five parts of COD for every
one part of TKN, the above ratio will be 0.026 to 0.2:1 based on a TKN:
MLVSS loading.
252
-------
•- PREDOMINATELY NO-
°- PREDOMINATELY NO,
50-
40-
^y
230-
H-
0
ro —
a <2o-
H
^
10-
O
1!
ii
1
o
o
.-
* - WASHOUT
o
INTENSIVE NITRITE FORMATION
\ WASHOUT
\ >y
i \ INTENSIVE NITRATE FORMATION
\
\
\.
\
\
\
" \
X •
^-~~.
— • f ,
50-
40-
K"
z
2 30-
t—
o
D_
t20-
t
z
^ 10-
i
U n_
\
o..
0
0
o
0
*>• _
* •«. •
•*
INTENSIVE NITRITE FORMATION
N. * WAS HOUT
^^^
^ INTENSIVE NITRATE FORMATION
•
0
s
A
/
0
400 3000
(FREE NH3/TVS)X IO"6
100 200 300 400 3000
(FREE NH5/TVS) X IO"6
FIGURE 93
NITRITE AND NITRATE FORMATION AS
AFFECTED BY THE FREE AMMONIA/TVS RATIO
-------
60
50-
40-
0 30.
o
£ 1 20-
z
I
0-
C
60-
50-
"• *
40-
z
0
3 30-
U_
p
• tr
z
o
10-
o
* o
•-PREDOMINATELY NO
°- PREDOMINATELY NOj
o o A -WASHOUT
O
O
o
• •
•
r-
" 1 I i i i i I i i i O i i i i i i i i i i
) 200 400 600 800 1000 0 200 400 600 800 10
(HN02/TVS)X IO"6
(HN02/TVS) X IO"6
FIGURE 94
NITRITE AND NITRATE FORMATION AS
AFFECTED BY THE NITROUS ACID/TVS RATIO
-------
In Figures 93 and 94, the active mass was estimated by the total volatile
solids concentration (TVS). It is recognized that this parameter is not
an accurate measure of the active mass but it was the closest approxima-
tion that could be made with the data. From Figure 93 it appears that
if the ratio of free NH3 to TVS was less than 0.000025, complete nitrate
formation could be accomplished. At higher ratios, nitrate formation
became inhibited and nitrite formation became predominant.
Free nitrous acid (Figure 94) did not seem to inhibit nitrate formation
probably because nitrites were not in high concentrations until after
the nitrate formation was inhibited.
Because of the limitation caused in pumping concentrated feed suspensions
into the laboratory units, higher loading factors than the ones tried in
the laboratory could not be used. A high loading factor of 2.1:1,
COD:MLVSS was tried at an SRT of 1 day (unit G1) and resulted in no
nitrification. This was likely due to the washout of the nitrifying
population from the units rather than the high loading.
Because TVS is an imperfect measure of active mass, it would be useful
to know the concentrations of free ammonia that existed in the units
when nitrate inhibition occurred. The actual free NH- concentrations in
the continuous nitrification units were compared to the percent nitrite
and nitrate production (Figure 95). The ratio used for comparison was
the ratio of % nitratification to the % nitrification (% HO^-H formed/
N02-N + NO,-N formed). This ratio represents nitratification as a frac-
tion of the overall nitrification that occurred under a specific set of
conditions. This approach was taken since free NH3 is toxic to the
nitrate formers and therefore should be directly related to this ratio.
Values of free NH3-N greater than 0.2 mg/1 were not included in Figure 95,
although they did occur in some units, to enable an expansion of the x
axis and the critical portion of the Figure. Above a free NhU-N concen-
tration of 0.2 mg/1, nitritification predominated. When the free NH^-N
concentration was greater than 0.02 mg/1, nitrate formation rapidly
decreased and nitrite formation increased. However, unit D, which had a
free NH,-N concentration of 0.033 mg/1 sustained complete nitratification.
This difference in the behavior of unit D may lie in the fact that it was
operated at an SRT of 216 days as compared to the other units which were
operated at lower SRT values, 9 to 18.6 days. The long SRT of unit D
might have provided an opportunity for the nitrifying population to adapt
to the slightly higher concentration of free NH3-N in unit D. The rela-
tively shorter SRT values, 9-18.6 days, are adequate to sustain a nitrifying
population to achieve optimum nitrification provided other conditions are
favorable for microbial growth. The increased tolerance of the nitrifying
population, although slight at the long SRT (216 days), may be explained
on the basis of microbial adaptation to higher concentrations of free NH3
rather than to the presence of a greater number of organisms.
255
-------
e- SRT - 216 DAYS
CONTINUOUS
NITRIFICATION
UNITS
SRT VALUES FOR THE
OTHER UNITS RANGED
FROM 9- 18.6 DAYS
FREE AMMONIA, mg/l
FIGURE 95
NITRATE FORMATION RELATED TO THE
FREE AMMONIA CONCENTRATION
256
-------
The continuous flow unit data are compared with the data from the batch
nitrification study in Table 38. It should be recalled that the batch
study data represent the ever changing situation in the batch units as
contrasted to the continuous flow unit data which were obtained under
equilibrium conditions. The residual ammonia concentration at which the
nitratification started to occur was determined from the data presented
in Table VI, Appendix. This was the ammonia concentration at which the
nitrites started to decrease with an increase in the nitrates. The per-
cent nitratification presented for the batch data should not be construed
as the true nitratification achieved. The percent values represent the
apparent nitratification only which is defined as the ratio of nitrate
concentration on any given day to the total nitrogen in the system on
that day [(N03 -N/TKN + N02-N + N03~N) xlOO].
Although the continuous flow unit and batch unit data were representative
of two different sets of conditions, certain observations can be made by
a comparison of the two. The residual ammonia concentrations at which
nitratification occurred in the batch units were within the range observed
in the continuous flow units, but the free NH--N concentrations in the
batch units were considerably higher than that in the continuous flow
units. This was due to the higher pH values of the batch units.
The concentrations at which nitratification occurred in the batch units
ranged between 0.036 to 0.11 mg/1 (Table 38). However, in the continuous
flow units, nitratification decreased rapidly above a free NH3-N concen-
tration of 0.02 mg/1 (Figure 95). The sustained nitratification at the
higher concentrations of free NH3-N in the batch units was presumably due
to the adaptation of the nitrifying organisms to those concentrations.
In the continuous flow units it appears that a high SRT is necessary to
achieve the higher tolerance of the nitrate forming organisms to free
NH3-N.
It is interesting that the nitrate formers were able to tolerate the very
high free NhL-N concentrations occurred on the first day of operation of
the batch units. These concentrations ranged between 22-192 mg/1 (Table
23). The fact that nitrification occurred in these units after prolonged
operation when low concentrations of free NH3~N were present suggests
that the nitrite and nitrate formers were in a dormant stage during the
early period of high free NH,-N. The organisms were not eliminated when
•J i
exposed at least temporarily to the higher free NH3-N concentrations.
The surviving nitrifying population became physiologically active when
the levels of the free NH3~N decreased to the range of 0.036-0.11 mg/1
and as a result nitrification occurred. ;
The respirometric experiment conducted with unit G1 mixed liquor showed
that the degree of nitritification decreased with increasing ammonia.
257
-------
TABLE 38. FREE NHg-N AND NITRATIFICATION IN BATCH AND CONTINUOUS FLOW UNITS
CONTINUOUS
Unit
A1
B1
PO C
en
oo
D
D1
E1
E1
F1
F'
FLOW UNITS
SRT (days)
10.6
9
11.6
216
12.5
18.6
10.3
13.5
12.2
Residual NH4-N
in unit (mg/1 )*
120
40
45
145
105
270
170
285
220
PH
5.8
5.8
5.6
5.6
5.6
4.85
5.75
5.6
6.2
Free NH3-N
(mg/1 )
0.02
0.013
0.01
0.033
0.02
0.01
0.05
0.053
0.162
% Nitrati-
fi cation
44
47.5
49.8
60.5
38
49.5
18.9
23.8
3
N03-N
N02-N + N03-N
0.87
0.95
0.95
1
0.99
0.99
0.44
0.58
0.06
-------
TABLE 38 concluded. FREE NHg AND NITRATIFICATION IN BATCH AND CONTINUOUS FLOW UNITS
ro
tn
BATCH STUDY
Unit
40
50
60
b
c
Maximum**
Residual
NH4-N
(mg/1 )
247
305
374
178
188
pH
8.5
8.5
8.55
8.4
8.35
Free
NH3-N
(mg/1 )
32
40
54
23
22
Residual***
NH4-N
(mg/1)
56
48
31
94
106
pH
6.25
6.2
6.3
6.15
6.25
Free
NH3-N
(mg/1 )
0.057
0.044
0.036
0.076
0.11
% Maximum
Apparent
Ni trati f i ca ti on****
65
63
59
55
50
**
NH.-N concentration under conditions of equilibrium
Ammonium concentration measured during the operation of the unit. This generally occurred on the
first day after starting the operation of the unit.
*** Represents the ammonium concentration when the nitrite concentration began to decrease and the
nitrate concentration began to increase indicating that nitratification was occurring.
**** Occurred at the end of the run, after both the maximum residual NH.-N and the residual NH.-N at
the onset of nitratification had decreased.
-------
However, even at a concentration of 2000 mg/1 of supplemental NH4-N,
nitrite production occurred indicating that nitritifying populations can
be quite tolerant of high NH^-N concentrations.
In the batch units, the tolerance of the nitrate formers to free ammonia
appears to have increased with an increase in the residual ammonia con-
centration at which nitratification occurred (Table 38). The free NH3-N
concentration at which nitratification occurred in units b and c was 0.076
and 0.11 mg/1 respectively. However, in units 40, 50, and 60, the nitra-
tification occurred at 0.057, 0.044, and 0.036 mg of free NH3-N/1 respec-
tively. Although the units 40, 50, and 60 received more concentrated
poultry manure suspensions and produced higher NH^-N and free NH3-N con-
centrations after one day of operation than units b and c, the nitrati-
fication did not occur until the free NH3 concentration was less than
that of units b and c. The high initial concentrations of NH^-N and pH
in the units 40, 50, and 60 were responsible for the volatilization of
NH3 which resulted in lower concentrations of free ammonia than those
noticed in units b and c. Thus the nitrifying population was adapted to
higher concentrations of free NH3-N in units fed with lower concentra-
tions of poultry manure suspensions than those fed with suspensions of
higher concentration.
Thus, it may be that as a result of the greater losses of NH.-N from
heavily loaded batch reactors, the tolerance of the nitrifying population
to free ammonia decreased. The free ammonia concentration at the onset
of nitrification in such reactors was found to be lower than the concen-
tration of free NH3-N in batch reactors where such ammonia nitrogen vola-
tilization did not take place.
In continuous flow units, the situation was different since equilibrium
conditions prevailed. The range of free NH3-N found under nitratifying
conditions was smaller in comparison to that of the batch units, i.e.,
0.01-0.033 compared to 0.036-0.057 mg/1 respectively (Table 38). Shock
loads of poultry manure are more likely to disrupt the nitrification process
in a continuous operation than in a batch process.
The relationship between the undissociated nitrous acid and the % nitriti-
fication/35 nitrification (N02-N)/(N02-N + N03-N) was presented in Figure 96.
Concentrations of undissociated HN02 were found to have no significant
effect on nitritification. From Figure 96, it can be seen that above a
concentration of 0.3 mg of free HN02/1, nitritification predominated.
The proportion of nitrite of the total oxidized nitrogen in the mixed
liquor varied with an increase in the undissociated HN02 concentration
as indicated by the wide band of (N02-N)/(N02-N + N03-N) values in Figure 96,
260
-------
i.o-
g
^
o
u.
i-
cc.
1—
z
1-
z
UJ
0
o:
UJ
a.
2
O
1—
O
u. .6-
^
z
h-
2 .4-
UJ
0
a:
UJ
0.
.2-
"
0-
V*" . . >x
T * GENERAL
• • BOUNDARY
1 CONDITIONS
• i /
1 /
1 • ^/
1 . — — —
1 "
1
1 *
1
1
1
1
. CONTINUOUS
1 NITRIFICATION
I UNITS
1
/
* •
i i i i i
012345
HN02 -mg/l
FIGURE 96
NITRITE FORMATION RELATED TO THE
NITROUS ACID CONCENTRATION
261
-------
Residual Ammonia - Although nitrification took place in the continuous
flow units, it is interesting that the mixed liquors had a high concen-
tration of residual ammonia which was found to increase with an increase
in the TKN loading (Figure 53). This is perhaps due to the inhibition
of the oxidation of ammonium to nitrite by the Nitrosomonas population
due to repression of enzyme synthesis by the end products (N0£ and NO^)
formed. This is likely the case because we have found that a nitrified
mixed liquor containing residual ammonia after denitrification could be
nitrified again. A plot of the NCL-N + NOg-N against the residual ammonia
concentration (Figure 97) shows a direct relationship, also suggesting
some inhibitory effect of nitrite and nitrate on ammonia oxidation. The
denitrification step decreased the inhibitory effect permitting the residual
ammonia to be nitrified. Thus in poultry waste it is difficult to achieve
a single step nitrification of all the ammonia present in the mixed liquor.
However, a nitrification-denitrification-nitrification sequence offers
promise in minimizing the NH»-N in the mixed liquor. Future research is
planned on the continuous cycling of nitrification and denitrification for
nitrogen control.
During the course of these laboratory studies only 50-60 percent of the
TKN could be nitrified and this was the only amount available for denitri-
fication. Percent nitrification can be reported in several ways. Many
investigators have chosen to express the nitrification as a percent of
the ammonia nitrogen converted to NO^-N and/or N03-N. In the presentation
of nitrification data with secondary effluents or where all of the TKN has
been converted to ammonia, this definition may be adequate. However, in
situations where all of the TKN has not been converted to NH.-N, expressing
nitrification in the above manner may be inadequate since further ammoni-
fication of the residual TKN and subsequent nitrification could occur. In
these studies with concentrated mixed liquors, nitrification data is pre-
sented in terms of the percent of the initial TKN nitrified.
In the treatment of municipal sewage, varying degrees of nitrification
were reported to be achieved by several investigators. In a nitrification
study of a primary effluent containing a TKN of 49.6 ± 4.6 mg/1, about 80%
was NH4-N (166). Approximately 95% nitrification of the TKN of this waste
occurred suggesting that a portion of the organic nitrogen was also nitri-
fied in addition to the available NH^-N. In contrast to these observa-
tions, only 50-60% oxidation of the TKN contained in a primary effluent
was reported (95). In another study, out of the 27.2 mg/1 of TKN con-
tained in a synthetic waste, only about 51-66% was nitrified (71). The
nitrification of raw sewage has been reported at various loadings (187).
In this study about 70% of the NH4~N present initially in the raw sewage
was nitrified. However, based on the TKN present in the raw sewage, and
the N03-N that appeared in the effluent approximately 35% of the TKN was
nitrified at lower organic loadings. In our studies the percent
262
-------
1200 n
x 1000 H
o>
e
z
I
-------
nitrification of the TKN was in the range of 50-60%. Assuming that
the residual ammonia nitrogen can be nitrified completely in a second
nitrification stage after the denitrification of the N02 + NOg-N formed
initially, we can obtain about an average of 75% nitrification of the
TKN originally contained in the poultry manure suspension. As noted
above, the percent of nitrification obtained in this study are com-
parable to those of other investigators when nitrification is defined
in terms of the TKN in the initial waste.
Although the exact nature of the remaining 25% of the TKN is not known,
it is likely that this is not available for deamination and subsequent
nitrification. Studies made about four decades ago with farmyard manure
showed that about 18-25% of the organic nitrogen of manure was present
in a form similar to the a- humus of soil and this was shown to be very
resistent to decomposition and further nitrification (188, 189). It
is also possible that some of the nitrogen assimilated by the microbial
population present in the chicken manure mixed liquors may be converted
into the resistant a- humus-like compounds. Besides this form of nitrogen,
microorganisms in general contain other resistant nitrogenous compounds
which also are present in poultry manure mixed liquor, since it contains
both living and dead microorganisms. In addition to these resistent
nitrogenous materials, feathers also contribute to the non-degradable
form of TKN. Thus the 75% of the TKN that could be nitrified represents
about the maximum degree of nitrification that can be achieved with
these poultry wastes.
The process of mineralization also depends on the C:N ratio of the
material undergoing transformation. If a material has a low C:N ratio
(rich in nitrogen) the tendency for the nitrogen to be mineralized is
greater. On the other hand, because of the increased supply of carbon
in proportion to nitrogen with materials having a high C:N ratio, micro-
bial synthesis may require all the available nitrogen. In the latter
case, no net mineralization will result and there will be increase in
both the amount of nitrogen temporarily incorporated in the protoplasm
of the microorganisms and the nitrogen incorporated into the hard-to-
degrade compounds.
Data on the mineralization of various materials having different C:N
ratios are reproduced here with a comparison of the mineralization
possible with chicken manure (Table 39).
The degree of mineralization obtained for the poultry manure in our
laboratory coupled with the potential of the mineralization of the
residual ammonia in the mixed liquor appears to be reasonable in view
of the low C:N ratio and is comparable to the mineralization possible
for the readily degradable nitrogenous materials such as casein, hoof,
and dried blood. Thus with poultry manure it may not be possible to
achieve a higher degree of mineralization than that noticed in our
studies.
264
-------
TABLE 39. ANALYSES OF MATERIALS AND PERCENTAGES OF NITROGEN MINERALIZED ON INCUBATION WITH SOIL
ro
CT>
in
Carbon
Material content %*
OTHER STUDIES (190)
Cellulose nitrate
Nitrated sawdust
Edestin (B.D.H.)
Hoof
Commercial casein
Formalized hoof
Casein
Deaminated casein
Dri ed bl ood (commerci al )
Formalized casein
Amy! nitrate
Crab shell meal
Chi tin
Lignin I - nitric acid complex(
Oat straw compost
Lignin II - nitrous acid complex
Lignin II - edestin complex
Lignin II - ammonia complex
Humic acid
Lignin I - ammonia complex
POULTRY MANURE
Dry chicken manure-
Dry chicken manure (194)
Fresh chicken manure (194)
27.8
28.7
50.8
50.0
44.0
50.8
50.6
47.3
54.0
47.7
45.2
10.5
47.0
54.3
41.8
50.7
61.2
64.9
58.6
63.0
33.9
-
-
Nitrogen
content %*
12.2
12.2
18.5
16.6
14.1
16.6
15.3
14.2
15.9
13.6
10.5
2.04
6.70
5.20
3.95
3.79
2.80
3.05
2.31
2.36
7.8
6.1
1.48T
C/N ratio
2.3
2.4
2.7
3.0
3.1
3.1
3.3
3.3
3.4
3.5
4.3
5.1
7.0
10.4
10.6
14.9
16.1
21.3
25.4
26.7
4.3
-
-
Nitrogen mineralized
(% of total -N)
Period of incubation (days)
40 80
1
5
76
71
73
54
75
67
60
56
1
52
44
23
5
21
-1
6
0
8
50-65
44b
40b
-2
10
80
75
80
71
82
74
65
73
0
65
49
30
11
27
-1
9
6
12
5<£
47^
* Moisture-free basis
t Wet basis
a_ This study
b_ 21 days incubation at 28°C
c_ 28 days incubation at 28°C
-------
Residual TKN - In the batch nitrification study, TKN in the range of
11-25% still remained in the units which were started at higher loadings
and operated for a prolonged time. The ammonia fraction of the TKN was
very small.
In one experiment, nitrified mixed liquor from a continuous flow unit
was denitrified. Subsequently, it was possible to nitrify the residual
ammonia of the denitrified mixed liquor. The renitrified mixed liquor
did not contain ammonia nitrogen. Assuming that all the continuous flow
units would behave in a similar fashion, a TKN of 11-38% will still
remain in them after the denitrification and a second step nitrification.
From our studies and the available information on the mineralization of
nitrogenous materials, it is unlikely that further mineralization will
occur of the residual TKN from sequential nitrification-denitrification
processes. Further studies are required to determine the nature of the
TKN that would be still present after such a sequence. The disposal of
such non-degradable and non-nitrifiable TKN on the land should not con-
tribute to the enrichment of water resources with nitrogen.
Effect of Dilution - Respirometric studies indicated that the oxygen
uptake and degree of nitrification were increased as a concentrated
nitrified mixed liquor was diluted. The data suggested that a solids
concentration of less than 1% favored higher oxygen uptake rates and
nitrification. However, from a practical standpoint, it may not be
beneficial to dilute concentrated poultry waste mixed liquors to accom-
plish this higher degree of nitrification, an increase of about 10-15%,
since this increase can be obtained in a nitrification step following
denitrification without any dilution.
Effect of SRT, Loading Factor and pH - The results of our study on the
effect of SRT on nitrification of poultry manure indicated that a mini-
mum of 2 days are required to sustain the process (Figure 98). These
results are in agreement with other data reported for municipal waste
(95, 96, 98, 166) and suggest that the growth kinetics obtained in
other studies are applicable to poultry manure as well.
The data from this study are compared with that published in other
studies in Figure 99. A strict comparison of this data is not possible
because the loading factor is expressed differently in each study.
Nevertheless, certain observations can be made. Beyond a certain loading
factor, nitrification will not occur. This may be due to free ammonia
inhibition at the higher loadings and to possible washout of the nitri-
fying organisms.
In these studies, a loading factor less than 0.8# COD/day/# MLVSS was
found to insure maximum nitrification. With the poultry wastes and the
experimental units of this study, it was not possible to obtain higher
loading factors and maintain an SRT to assure optimum nitrification.
Unlike the inhibitory effect of low pH on nitrification reported for
municipal sewage (126, 140, 176), units fed with poultry manure maintained
266
-------
60-|
Q
UJ
u.
EE
t 40
ui
20-
I.
o
.jo
CONTINUOUS FLOV/
NITRIFICATION UNITS
I I I I I I I I I I I I
5 10
i i
15
i i \ i
20
SOLIDS RETENTION TIME ( DAYS)
FIGURE 98
NITRIFICATION RELATED TO
SOLIDS RETENTION TIME
267
-------
AUTHOR
SYMBOL LOADING FACTOR
r\3
en
00
100
0 80
Ld
Lu
CC
h;
2 60
40-
LJ
O
cn
20-
0
0.4
0.8 12
LOADING
MECHALAS, ET AL
BALAKRISHNAN
DOMESTIC WASTE
SYNTHETIC WASTE
WUHRMANN
1.0 MG/L D.O.
7.0 MG/L D.O.
THIS STUDY
LB. COD /DAY
L8. MLVS
LB. BOD/DAY
LB. MLSS
LB. BOD/DAY
LB.. MLVS
LB. COD/DAY
LB. MLVSS
£ - NGRATIFICATION
O - NITRITIFICATION
1.6
FACTOR
2.0
2.4
FIGURE 99
NITRIFICATION AS AFFECTED BY THE LOADING FACTOR -
COMPARISON OF DATA FROM VARIOUS STUDIES
-------
nitrification even at a pH of 4.9. Adjusting the pH to neutrality or to
slightly alkaline conditions did not cause any significant increase in
the degree of nitrification. Nitrification was definitely inhibited at
pH 4. At pH 10 and 11 it was inhibited in dilute poultry manure sus-
pensions. However, the mixed liquor of an oxidation ditch containing
a high solids concentration showed signs of inhibition initially but
recovered in approximately 30 and 70 hours respectively at these two pH
values. The results of this study indicated that there is no apparent
need to control the pH in the aerobic systems used for the nitrification
of poultry manure.
Denitrification
General - The denitrification of the nitrified mixed liquor can be achieved
by the metabolic activity of the facultative heterotrophs under anoxic
conditions. Their growth under such conditions is dependent on the
availability of a carbon source at the expense of nitrite or nitrate or
both as electron acceptors. Batch studies in the laboratory yielded
information on the nature of the denitrification process as applied to
nitrified poultry manure suspensions. The following is a discussion
pertaining to some of the important factors that govern the process.
Anaerobiosis - The maintenance of anoxic conditions is necessary to
achieve denitrification. It was not possible to denitrify the nitrified
mixed liquor developed in aerobic systems fed with dilute poultry manure
suspensions without ensuring a lack of dissolved oxygen and avoiding
oxygen input. The cessation of aerating the mixed liquor alone did not
bring about denitrification due to the endogenous respiration of the
microbial mass. Perhaps some denitrification would have resulted after
a long period of time but such an effort was abandoned because the results
of such an experiment would have no intrinsic practical value. However,
under strict anoxic conditions, about 25-35% of the oxidized nitrogen
was lost within 3-6 hours from dilute nitrified mixed liquors with little
loss in the subsequent 18 hour period.
Plateau in Denitrification - In most of the denitrification runs there
was a rapid loss of nitrogen in the first few hours accompanied by a
plateau for several hours. This plateau was once more followed by a
rapid loss of nitrogen. The biphasic pattern was observed with both
N02-N and NOg-N removal. Such a plateau seemed to occur in the denitri-
fication of municipal waste (166). Although such a plateau could be
visualized in their data, these authors represented the trend by a linear
plot (Figure 100). The probable reason for the causation of the plateau
was explained under denitrification Run VI. Further study is needed to
determine whether such a plateau is observed primarily in the denitri-
fication of poultry manure slurries or is prevalent in all nitrified
systems fed with different wastes.
Denitrification Rates - A compendium of the denitrification rates obtained
in this study under various experimental conditions and the ones reported
269
-------
50-i
i°
o - UNIT 1
• - UNIT 2
AUTHORS LINE OF BEST FIT
OUR ESTIMATE OF BEST FIT
0
6 8
TIME (hours)
10
FIGURE 100
DENITRIFICATION DATA FROM REF- 166
INTERPRETATION OF RESULTS
270
-------
for municipal wastes by others is given in Table 40. There is a wide
range in the rates of denitrification. The variation in the denitri-
fication rates can be due to: a) the different types of hydrogen dona-
ting material (endogenous or exogenous), b) the parameters used to
estimate the hydrogen donors and the active mass of organisms, c) dif-
ferent time intervals used in computing the rates, and d) different
experimental procedures. In this table the denitrification rate was
computed as the amount of oxidized nitrogen that was lost per unit time
per gram of solids. These solids were used as an estimate of the
hydrogen donating substances and the active mass.
All the denitrification rates except those for Runs I and II were
average rates computed over a period which included the time during
which the plateau occurred. It is likely that the plateau in Runs I
and II would have been observed if the experiments had run for a longer
period. The rates of denitrification of these units reflect the maximum
rate. In most of the denitrification runs, high rates were observed
during the first few hours, and were followed by a plateau. By inclu-
ding the period during which the plateau occurred in computing the rates,
the overall rate of denitrification was substantially reduced.
Our study indicated that the rates of NOp-N removal were higher than
those for NO^-N removal (Run V) (Run IX). In municipal waste high con-
centrations of NOp-N are seldom encountered and there is little infor-
mation on its removal. With an agricultural wastewater, however, it was
reported that the NOp-N formed as an intermediate in the denitrification
of N03-N was not denitrified readily (63).
Denitritification rates were higher than denitratification rates in
nitrified poultry manure. These studies indicated that at a loading
factor greater than 0.15# COD/day/# MLVSS, nitrites predominated (Figure
99). In addition, high concentrations of NOp-N with relatively low N03-N
could be achieved at loadings up to at least 0.8. By practicing nitriti-
fication, in addition to the high loadings that are possible, there is
the trade off of relatively faster removal of nitrogen than one would
anticipate in a highly nitratifying system.
The denitritification, denitratification, and the overall denitrification
rates observed with poultry manure slurries are within the range of those
reported by others (Table 40) who used the endogenous oxygen demand of
the denitrifying system. Higher denitrification rates were obtained
when the systems were supplemented with glucose or poultry manure.
Higher rates occurred with glucose. The addition of poultry manure
increased the TKN of the resultant mixed liquor. To enhance the rate
of denitrification, supplementation of raw chicken manure as a hydrogen
donor is not advantageous since a greater TKN results. The data showed
that one can expect complete denitrification in a reasonable time period
(4-6 days) with mixed liquors containing a high (1-6%) concentration of
271
-------
TABLE 40. OBSERVED RATES OF DENJTRIFICATION
Study
Solids
concentration
-(mg/1)
Rate of
denitrifi cation
mg-N/hr/gm solids
Time used in
computing rate
Reference and
comments
ro
RESULT OF OTHER INVESTIGATIONS
Batch
Batch
Batch and
continuous
Batch
THIS STUDY
Run I
Run II
25°C
29°C
18°C
29°C
18°C
20°C
20°C
20°C
35°C
5356 (MLSS)
4630 ( " )
4630 ( " .,«•)
3670 ( " " )
3670 ( " )
3670 ( " )
4143 ( " .)
4143 ( " )
4143 ( " )
6300 (MLVS)
6300 ( " )
6300 (MLVS)
6300 ( " )
3044(MLSS)
5350-
2680 (Total Solids)
2280 ( " " )
'2280 (•'•" " )
u 1'9
u 2.5
2.4
1.8
1.6
1.6
1.5
1.4
1.4
0.39-0.89 (ave 0.58)
0.21-0.34 (ave 0.26)
0.68 (average)
0.3 (average)
0.55
1.09
4.5iT
5-7F
7.4^-
l_hr
1 hr
2 hr
1 hr
2 hr
3 hr
1 hr
2 hr
3 hr
6-10 hr
7.5-24 hr
11 hr
3 hr
6 hr
3 hr
3 hr
95
166
166
71
-------
TABLE 40 continued. OBSERVED RATES OF DENITRIFICATION
—i
CO
Study
Run III
Run IV
Run V
20°C
35°C
20°C
35°C
20°C
35°C
Solids
concentration
(mg/1 )
185(£ (MLSS)
1850- ( " )
4775T (MLSS)
4755- ( " )
5300 (MLSS)
5300 (MLSS)
Rate of
denitrification
mg-N/hr/gm solids
0.8
1.6
0.4
0.6
0.4-
0.18—
O.'82f-
0.16L
Time used in
computing rate
3 days
2 days
3 days
3 days
6 days
1 day
3 days
3 days
Reference and
comments
high N03-N incom-
plete denitrification
high NO,-N, low
NH--N *•
almost complete
denitrification
Run VI 20°C
1.
2.
3.
Run VII 20°C
1.
2.
1020 washed MLSS
+ 100 mg/1
glucose
0.7
7 days
MLSS + 100 mg/1
glucose
1020 MLSS
4790 MLSS
4740 MLSS
+ 1 .726 gm of
glucose/1
0.64
0.64
0.067
0.093
7 days
7 days
7 days
13 days
high N02-N,
essentially dem'tri-
tification, complete
denitritification in
1 and 2, partial de-
nitritifi cation in 3
partial denitrifi-
cation in 1 and 2
and almost complete
denitrification in
3, 4, and 5
-------
TABLE 40 continued. OBSERVED RATES OF DENITRIFICATION
Study
3.
Solids
concentration
(mg/1)
4830 MLSS
+ 2.3452 gm
glucose/1
Rate of
denitrifi cation
mg-N/hr/gm solids
0.32
Time used in
computing rate
7 days
Reference and
comments
PO
*-J
45.
Run X
4. 8330 MLSS
+ 25.7 gm
poultry manure/1
5. 10740 MLSS
+ 51.4 gm
poultry manure/1
20°C 5220-51,400
0.15
0.16
0.17-0.3
7 days
7 days
3-14 days
Very high N02-N and
low NO--N oxidation
ditch mixed liquor.
Increased rate of
denitritifi cation
was observed due to
N02 +
NO, supplemen-
tation with no effect
on denitratification
rate
-------
TABLE 40 concluded. OBSERVED RATES OF DENITRIFICATION
Study
Solids
concentration
(mg/1)
Rate of
denitrification
mg-N/hr/gm solids
Time used in
computing rate
Reference and
comments *
ro
~^i
en
Run XI 20°C 56,930 (MLSS)
initial pH adjusted
to 4 to 11
0.047-0.38
4-16 days
Very high N02 and
low
NO, oxidation
ditch mixed 1iquor.
Both the denitriti-
fication and denitra-
tification rates
increased up to pH 8
a^ Supplemental carbon included
b_ Rate prior to plateau
c_ Suspended solids include solids from supplemented poultry manure
d_ No supplemental hydrogen donors
e_ Denitritification
f Denitratification
-------
solids, such as the oxidation ditch mixed liquor, without supplemen-
tation of any hydrogen donors.
If high rates of denitrification are desired, it is necessary to supply
a readily available hydrogen donor to poultry waste treatment systems.
These donors should not increase the nitrogen content of the system.
However, high rates of denitrification may not be essential in operating
denitrification systems with poultry wastes and endogenous hydrogen
donors in relatively concentrated mixed liquors may be sufficient.
Nitrogen and COD Removal - In a highly nitritified system, the observed
COD removal that occurred for a given N02-N removal was in close agree-
ment with the COD removal predicted by the stoichiometric relationship
described by McCarty et.al. (63). However, in a highly nitratified
system the predicted COD removal did not agree with the observed COD
removal. The COD removals in the nitratified systems were higher than
the predicted COD removal. Enough qualitative data on the nature of
various mixed liquors that were subjected to denitrification was not
available to explain the quantitative differences observed in the
removal of COD and NO-j-N. Nevertheless, since the stoichiometric rela-
tionships are seemingly applicable to systems containing high concen-
trations of NOp-N, and since highly nitritifying systems are only
encountered at high chicken manure loadings, one should be able to pre-
dict by determining the COD of the mixed liquor whether all the N02-N
can be denitrified or not.
pH and Temperature - These studies indicated that pH control was not
essential to accomplish denitrification. Although extreme pH conditions
do affect the denitrification adversely, the pH that resulted in the
nitrification units (5.0-6.5) will not be detrimental and the control
of pH will not be necessary in the ensuing denitrification stage. The
maximum rates of denitrification were observed in the unit adjusted to
an initial pH of 8. The reactors in which the pH was adjusted initially
to 6, 8, 10, and 11 also denitrified completely with the pH 10 and 11
systems taking longer time to denitrify than the others. In all the
reactors the pH shifted towards 8.5. An increase in the temperature
from 20 to 35°C increased the rate of denitrification. In a full scale
system it may not be advantageous to raise the temperature to accomplish
higher denitrification rates.
Gases Produced - The results of the semi-quantitative gas chromatographic
analysis showed that N2 was the major end product of denitrification. A
measureable amount of N20 also was produced. Other gases produced inclu-
ded H2, C02, NO, and a gas tentatively identified as CH.. The implication
of the N20 produced as an atmospheric pollutant and its effect on animal
health was not investigated. It may be a problem if produced in large
quantities and if there is inadequate ventilation in the animal confinement
area.
276
-------
Degree of Nitrogen Removal Possible in Poultry Manure - The amount of
nitrogen that can be removed by a microbial nitrification-denitrification
process depends primarily on the amount of nitrogen that can be oxidized
in the nitrification step and on the amount of hydrogen donors that are
readily available in the denitrification step. This study showed that
a significant amount of residual ammonia was left in the mixed liquor
even at its maximum nitrification and at high SRT values. This residual
ammonia can be nitrified effectively after the mixed liquor is denitrified.
It should be possible to remove the N02~N and NCL-N formed in the second
nitrification step by subjecting the mixed liquor to denitrification again
if enough readily available hydrogen donors are left in the system. In
poultry manure mixed liquors the availability of hydrogen donors should
not be a problem since they have a high COD. Assuming that it is possible
to denitrify the oxidized nitrogen in a second denitrification step,
about 75-80% nitrogen removal can be expected based on the computations
given in Table 41. As can be seen from these computations, about 8.5%
of feed TKN was removed not directly due to the denitrification of the
nitrified mixed liquor. This difference could be due to a sampling and
experimental error.
ENGINEERING SIGNIFICANCE OF THE RESEARCH
There is a considerable lack of information available on the feasibility
of biological methods for the removal of nitrogen from highly nitrogenous
wastes such as poultry manure. This research was undertaken to explore
whether nitrogen can be removed from poultry manure utilizing a microbial
nitrification-denitrification scheme. Besides exploring the biochemical
and microbiological feasibility of the process, the effect of several
parameters were also studied to suggest design criteria for a nitrogen
removal system.
Seed and Air Supply - Since nitrification is brought about by a few
specialized autotrophs which are aerobic, it is essential that nitri-
fication units be started with an adequate amount of seed and aerated
at a rate to ensure at least a few mg/1 of dissolved oxygen in the mixed
liquor. A seed concentration of 1% by volume of a well settled nitrified
mixed liquor solids, or 10% by volume of the nitrified mixed liquor was
found to induce active nitrification in laboratory units. Aerating
chicken manure alone will not induce nitrification and seeding is essential,
Loading and SRT - At loading factors less than 0.8# COD/day/# MLVSS,
nitrification occurred at the maximum permitted by the conditions in the
system. At loading rates less than 0.15, nitrate was the primary end
product. Predominance of NO?-N can be expected at loading factors
greater than 0.15.
A minimum SRT of 2 days has to be maintained to sustain the process of
nitrification. Even at very high SRT values, it is difficult to nitrify
to nitrates at higher loading rates. Since it is easier to denitrify N02-N
277
-------
TABLE 41. NITROGEN REMOVAL POSSIBLE IN A PROCESS INVOLVING
TWO SEQUENCES OF NITRIFICATION AND DENITRIFICATION
The mixed liquor sample taken for denitrifi cation was taken on July 8,
1970 from Unit A1 operating at an SRT of about 10 days. The denitri-
fication was started on July 8, 1970. The results of this experiment
have been reported in the section on "Nitrification of the Denitrified
Manure".
1 . Average Feed TKN 320 mg/1
2. Conditions after first step nitrification
TKN in mixed liquor 156 mg/1
N03-N mixed liquor 190 mg/1
NH4-N mixed liquor 88 mg/1
% nitrification 59.3
% nitrogen balance +8.1
3. Conditions after first step denitrification
TKN 126 mg/1
N03-N 0 mg/1
NH4-N 25 mg/1
4. Conditions after second step nitrification
TKN 63 mg/1
N03-N 40 mg/1
NH4-N 0 mg/1
% nitrification 31 .8
% nitrogen balance (between 3 and 4) -18.7
5. % overall nitrification = ^n = 71.8
[between 1 and (2 plus 4)] ^u
6. Assuming that the N03-N formed in step 4 can be denitrified, the %
total nitrogen that can be removed will be:
32°32063 x 100 = 80.3
7. % nitrogen removal not accounted due to possible denitrification of
oxidized nitrogen formed in 2 and 4 above:
80.3 - 71.8 = 8.5
Difference is within the limits of sampling and experimental error.
278
-------
than N03-N, a pragmatic approach for the denitrification of poultry manure
will be to design for such a highly nitritifying system. By designing a
system for nitritification, the following advantages may be obtained: a)
use of high loading factors thereby reducing the volume of the units
and b) formation of N02-N which can be denitrified more rapidly than N03-N.
Based upon these studies, it appears that a loading factor in the range
of 0.6-0.8# COD/day/# MLVSS (Figure 99) will produce the desired advan-
tages. We were unable to study the effect of higher loading factors.
It may be possible that even higher loading factors will yield satis-
factory results provided that the SRT of the system is adequate for
the maintenance of the nitrifying population.
The formation of nitrate was hindered by the presence of free ammonia
which was more inhibitory than undissociated nitrous acid to the nitrate
forming bacteria. Since higher loadings of chicken manure result in
increasing concentrations of N02-N and low pH, undissociated nitrous
acid will be present in larger quantities as the concentration of NCL-N
increases. As the N02~N concentration increased in the system, the
residual NH^-N concentration also increased. This suggests that high
concentrations of N02-N also are inhibitory to the oxidation of NH.-N.
The residual NH.-N can be nitrified in a second nitrification step after
the N02-N and NO^-N are denitrified. High N02-N and N03-N concentrations
can be detrimental to NH.-N oxidation. To achieve a very high removal
of nitrogen, an additional nitrification step for the conversion
of the residual NH.-N to N02~N or N03-N followed by another denitrifi-
cation step to remove the oxidized nitrogen forms appears to be essential.
Without the second nitrification step it was possible to remove 50-60%
of the initial TKN. With this second step followed by denitrification
it was possible to remove a total of 65-75% of the initial TKN. In the
second denitrification step, it is likely that the COD of the mixed
liquor is adequate to bring about the denitrification.
pH and Temperature - The studies indicated that there is no need to
control the pH in either the nitrification or denitrification stages.
Although the rates of these processes can be increased by adjusting the
pH slightly to the alkaline side of neutrality, the results accomplished
by such adjustments are not significant enough to warrant the recommen-
dation of a pH control mechanism in an actual process.
Low temperatures could have a detrimental effect. If aerobic systems
such as environmentally controlled indoor oxidation ditches are used for
the handling and treatment of poultry manure, the detrimental effect of
low temperature on the processes of nitrification and denitrification
can be avoided.
279
-------
Summary - This study has shown that is is technically feasible to incor-
porate a nitrification-denitrification sequence with the biological treat-
ment of animal wastes to control the nitrogen content of these wastes.
APPLICATION OF NITRIFICATION-DENITRIFICATION SEQUENCE TO OXIDATION DITCH
OPERATION
Although this research indicated that the nitrogen removal from poultry
manure can be accomplished by the application of the principles of con-
ventional biological waste treatment, e.g., a modified activated sludge
process with nitrification followed by denitrification, the sophisticated
operations of such a process may be an additional burden as well as an
item that would need constant attention by the farmer. However, the
following observations made with an in-house oxidation ditch treating
poultry wastes which was in operation at the Animal Waste Management
Laboratory offer promise for the application of a nitrification-denitri-
fi cation sequence to this process for the removal of nitrogen: a)
nitrification was achieved in an oxidation ditch after starting it
properly, b) nitrification was sustained with constant manure input
from the chickens up to at least a solids concentration of about 8%
total solids, c) denitrification was accomplished within a week to two
weeks at these high solids concentration without producing odorous con-
ditions and at the savings of power, and d) nitrification was reestab-
lished without difficulty after diluting the ditch contents and restart-
ing the ditch.
COST CONSIDERATIONS
The current study was undertaken exclusively to explore the biochemical
and microbiological feasibility of removing nitrogen from poultry manure
using a nitrification and denitrification sequence. As such, no study
on estimating the cost of the process was made. It is contemplated that
when future pilot plant studies are completed, an estimate of the cost
involved in operating the process can be made.
280
-------
ACKNOWLEDGEMENTS
This research was supported by the Environmental Protection Agency under
project numbers 13040 DDG and 13040 DPA, and by the College of Agriculture
and Life Sciences of Cornell University, Agricultural Engineering Department
The guidance of Mr. Jeffery D. Denit, who served as Project Officer is
gratefully acknowledged. Mr. Harold Bernard, Mr. M. R. Scalf, and
Mr. Lee Mulkey provided significant assistance in the direction of the
project and in the review and completion of this report.
The research performed by Miss A. Amisano, Mr. L. Deustua, Mrs. S. Eswara,
Miss N. Gates, Mrs. A. Rao, and a number of part time technicians con-
tributed significantly to the completion and success of the project.
Technicians in the Department of Agricultural Engineering were of con-
siderable help in the construction and preparation of the research and
pilot plant equipment.
Dr. A. J. Francis of the Department of Agronomy planned and conducted the
gas chromatographic analyses of the denitrification gases. His assistance
and cooperation were deeply appreciated.
The encouragement and support by the Department of Agricultural Engineering,
the Department of Poultry Science, and the Director of Research of the
College of Agriculture and Life Sciences, a statutory unit of the State
University, was gratefully appreciated.
Finally, the patience and skill of Mrs. Colleen Raymond in typing and
assembling the final manuscript are most sincerely appreciated.
281
-------
REFERENCES
PHOSPHORUS REMOVAL
1. Owen, R., "Removal of Phosphorus from Sewage Plant Effluent with Lime",
Sew. and_ IncL Wastes 25, 548-556, 1953.
2. Rudolfs, W., "Phosphates in Sewage and Sludge Treatment", Sew. Works
Jour. 19, 43-47, 1947.
3. Gulp, R.L., "Wastewater Reclamation by Tertiary Treatment", J_. Water
Poll . Cont. Fed. 35_, 799-806, 1963.
4. Lea, W.L., Rohlich, G.A., and Katz, W.J., "Removal of Phosphates from
Treated Sewage", Sew, and^ _Ind_. Wastes 26, 261-275, 1954.
5. Barth, E.F. and Ettinger, M.B., "Mineral Controlled Phosphorus Removal
in the Activated Sludge Process", J. Water Poll. Cont. Fed. 39, 1362-
1374, 1967.
6. Schmid, L. and McKinney, R.E., "Phosphate Removal by a Lime Biological
Treatment Scheme", J_. Water Poll. Cont. Fed. 41_, 1259-1276, 1969.
7. Buzzell, J.C. and Sawyer, C.N., "Removal of Algae Nutrients from Raw
Sewage with Lime", J_. Water Poll. Cont. Fed. 39_, R16-23, 1967.
8. Seiden, L. and Patel, K., "Mathematical Model of Tertiary Treatment
by Lime Addition", Final Report, Contract 14-12-416, Federal Water
Pollution Control Administration, USDI, Cincinnati, Ohio, 1969.
9. Dunseth, M.G., Salutsky, M.L., Ries, K.M. and Shapiro, J.J., "Ultimate
Disposal of Phosphate from Waste Water by Recovery as Fertilizer",
Contract 14-12-171, Program 17070 ESJ, Federal Water Quality Admin-
istration, USDI, Cincinnati, Ohio, 1970.
10. Recht, H.L. and Ghassemi, M., "Kinetics and Mechanisms of Precipita-
tion and Nature of the Precipitate Obtained in Phosphate Removal
from Wastewater Using Aluminum (III) and Iron (III) Salts", Contract
14-12-158, Program 17010 EKI, Federal Water Quality Administration,
USDI, Cincinnati, Ohio, 1970.
11. Loehr, R.C. and Johanson, K., "Removal of Phosphate from Duck Waste-
waters - 1970 Laboratory Study", Contribution 71-2, Agricultural
Waste Management Program, Cornell University, Ithaca, New York, 1971.
12. Standard Methods for the Examination, of Water and Wastewater, 12th Ed.,
American Public Health Association, New York, 1965.
283
-------
13. Menzel, D.W. and Corwin, N., "The Measurement of Total Phosphorus in
Seawater Based on the Liberation of Organically Bound Fractions by
Persulfate Oxidation", Limn, and Ocean. TO, 280-282, 1965.
14. Theriault, E.J. and Clark, W.M., "An Experiment Study of the Relation
of Hydrogen Ion Concentrations to the Formation of Floe in Alum
Solutions", Public Health Reports 36. 181-198, 1923.
15. Loehr, R.C., "Animal Wastes - A National Problem", JSED-ASCE 95 SA2,
189-221, 1969.
AMMONIA DESORPTION
16. Mellor, J.W., A Comprehensive Treatise on Inorganic and Theoretical
Chemistry Vol VIII, Longmans, Green and Co., London, 1928.
17. Holmyard, E.J., Makers of Chemistry. Oxford-Clarendon Press, Oxford,
1931.
18. Liebig, Justus von, Chemistry in its Application to Agriculture and
Physiology, Revised and Enlarged Edition of the Fourth London Edition,
Wiley & Putnam, New York, 1847.
19. Russell, E.J., Soil Conditions and Plant Growth, Eighth Edition, Long-
mans, Green and Co., London, 1952.
20. Adams, S.H., Modern Sewage Disposal and Hygienics. E. & F.N. Spon, Ltd.,
London, 1930.
21. Rudolfs, W. and Chamberlin, N., "Loss of Ammonia Nitrogen from Trick-
ling Filters", Indust. and Engin. Chem. 23_, 828-830, 1931.
22. Blain, W.A., "Discussion - Ammonia Nitrogen Losses from Streams", JSED-
ASCE 95 SA5, 974-978, 1969.
23. Melamed, A. and Saliterm'k, C., "Removal of Nitrogen by Ammonia Emission
from Water Surfaces", Developments in Water Quality Research: Pro-
ceedings of the Jerusalem International Conference on Water Quality
and Pollution Research. June 1969. Edited by H.I. Shuval, Ann Arbor-
Humphrey Science Publishers, Ann Arbor-London, 165-172, 1970.
24. Nesselson, E.J., "Removal of Inorganic Nitrogen from Sewage Effluent",
Ph.D. Thesis, University of Wisconsin, Madison, Wisconsin, 1953.
25. Kuhn, P.A., "Removal of Ammonium Nitrogen from Sewage Effluent", M.S.
Thesis, University of Wisconsin, Madison, Wisconsin, 1956.
284
-------
26. Gulp, G. and Slechta, A., "Nitrogen Removal from Waste Effluents",
Public Works 97_. 90-93, February 1966.
27. Handbook of Chemistry and Physics, 47th Edition, Chemical Rubber Co.,
Ohio, 1966-1967.
28. Norman, W.S. and Hu, T.T., "Gas Absorption", Chemical Engineering
Practice, Vol 6, Fluid Systems II, Edited by H.W. Cremer & T. Davies,
Butterworths Scientific Publications, London, 1-73, 1958.
29. Lewis, W.K. and Whitman, W.G., "Principles of Gas Absorption", Indust.
and Engin. Chem. 1_6_, 1215, 1924.
30. Baines, S., "Some Aspects of the Disposal and Utilization of Farm
Wastes", J_. Proc. Inst. o£ Sew_. Purif., 578-588, 1964.
31. Water Pollution Research-!963, Report of the Water Pollution Research
Board with the Report of the Director of the Water Pollution Research
Laboratory, (HMSO, London), 73-77, 1964.
32. Loehr, R.C., "Treatment of Wastes from Beef Cattle Feedlots - Field
Results", Proc. Conf. on Animal Waste Management, Cornell University,
Ithaca, New York, 225-241, 1969.
33. Miner, J.R., "Water Pollution Potential of Cattle Feedlot Runoff",
Ph.D. Thesis, Kansas State University, Manhattan, Kansas, 1967.
34. Bayley, R.W., "Desorption of Waste Water Gases in Air", Parts 1 and 2,
Effluent and Water Treatment Journal (London), 78-84; 115-161, 1967.
35. Jeris, J.S., "A Rapid COD Test", Water and Wastes Engineering 4, 89-91,
1967.
36. Harkins, W.D., and Jordan, H.F., "A Method for the Determination of
Surface Tension and Interfacial Tension from the Maximum Pull on a
Ring", J_. Am_. Chem. Soc., 52^, 1752-1772, 1930.
37. Harkins, W.D., "Determination of Surface Tension and Interfacial
Surface Tension", Physical Methods of Organic Chemistry, Vol I,
Edited by A. Weissberger (Interscience Publishers, N.Y.), 149-201,
1945.
38. Mark, H., "Determination of Viscosity", Physical Methods of Organic
Chemistry, Vol I, Edited by A. Weissberger (Interscience Publishers,
N.Y.j, 135-147, 1945..
39. Brookfield "Synchro-lectric" Viscometer Instruction Manual, Brookfield
Engineering Laboratories, Stoughton, Massachusetts, 1-12.
40. Symons, J.M., "Simple, Continuous Flow, Low and Variable Rate Pumps",
J. Water Poll. Cont. Fed. 35, 1480-1485, 1963.
285
-------
41. Downing, A.L., "Aeration in Relation to Water Treatment", Proc. Soc.
Water Treatment and Examination 7. part 2, 66-101, 1958.
42. Hater Pollution Research - 1962, The Report of the Water Pollution
Research Board with the Report of the Director of the Water Pollution
Research Laboratory, (HMSO, London), 12-13, 1963.
43. Esvelt, L.A. and Hart, H.H., "Treatment of Fruit Processing Wastes by
Aeration", J_. Water Poll. Cont. Fed. 42, 1305-1326, 1970.
44. Zanoni, A.E., "Secondary Effluent Deoxygenation at Different Tempera-
tures", J_. Water Pol 1. Cont. Fed. 4J_, 640-659, 1969.
45. Gotaas, H.B., "The Effect of Temperature on BOD Reaction Rates",
Sew. Works J. 20., 441-449, 1948.
46. Aiba, S., Humphrey, A.E. and Millis, N.F., Biochemical Engineering.
Academic Press, 1965.
47. Aiba, S. and Huang, S.Y., "Some Considerations on the Membrane-
Covered-Electrode", J_. Ferment. Techno!. 4_7_, 372-381, 1969.
48. Prather, B.V., "Wastewater Aeration May Be Key to More Efficient
Removal of Impurities", Oil and Gas Journal 57, 78-89, November 30,
1959.
49. Prather, B.V., "Chemical Oxidation of Petroleum Refinery Wastes",
Paper presented at 13th Industrial Wastes Conference, Oklahoma State
University, Stillwater, Oklahoma, November 1962.
50. Prather, B.V- and Gaudy, A.F., "Combined Chemical, Physical, and Bio-
logical Processes in Refinery Wastewater Purification", Paper pre-
sented at the 29th Midyear meeting of the American Petroleum Insti-
tutes, Div. of Refining, St. Louis, Missouri, May 1964.
51. Bennet, G.E., "Development of a Pilot Plant to Demonstrate Removal of
Carbonaceous, Nitrogenous, and Phosphorus Materials from Anaerobic
Supernatant and Related Process Streams", Report of Project 17010 FKA,
FWQA, Dept. of Interior, 1970.
NITRIFICATION-DENITRIFICATION
52. Loehr, R.C., "Pollution Implications of Animal Wastes - A Forward
Oriented Review", USDI-FWPCA, Robert S. Kerr Water Research Center,
Ada, Oklahoma, 1968.
53. Midgley, A.R. and Dunklee, D.E., "Fertility Losses from Manure Spread
During the Winter", Bulletin #523, Agr. Exp. Station, Univ. of Vermont,
1945.
286
-------
54. Minshall, N.E., Witzel, S.A., and Nichols, M.S., "Stream Enrichment
from Farm Wastes", JSED-ASCE 96, SA 5, 513-524, 1970
55. Stewart, B.A., Viets, F.G., Hutchinson, G.L., "Agriculture's Effects
on Nitrate Pollution of Ground Water", J. Soil and Water Conservation,
23_, 13-15, 1968.
56. Loehr, R.C., "Effluent Quality from Anaerobic Lagoons Treating Feedlot
Waste", JWPCF 39, 384-391, 1967.
57. Loehr, R.C., "Treatment of Wastes from Beef Cattle Feedlots - Field
Results", Cornell Agricultural Waste Management Conference, 225-241,
1969.
58. Koelliker, J.K. and Miner, J.R., "Use of Soil to Treat Anaerobic
Lagoon Effluent - Renovation as a Function of Depth and Application
Rate", Paper presented at the ASAE Meeting, Lafayette, Ind., 1969.
59. Fitzgerald, G.P., "Stripping Effluents of Nutrients by Biological
Means", in Algae and Metropolitan Wastes, R.A. Taft San. Engr. Center,
Rept. W61-3, 1961.
60. Dugan, G.L., Golueke, C.G., Oswald, W.J. and Mindell, C.R., "Photo-
synthetic Reclamation of Agricultural and Solid and Liquid Wastes",
Report 69-5, Sanitary Engineering Research Laboratory, Univ. of
Calif., Berkeley, 1969.
61. Haltrich, W. and Jager, B., "Beobachtungen bei der Biologischen
Reinigung Nitrathaltiger Industriller Abwasser mit Denitrifizierendem
Belebtschlamm", GWF 104, 347-352, 1963.
62. Bringmann, G. and Kuhn, R., "Gesteverte Nitrat-Eliminierung aus Ober-
flachenwasser durch Wasserstaffoxidation", Gesundheitsingineur 84,
213-219, 1963.
63. McCarty, P.L., Beck, L. and St. Amant, P., "Biological Denitrification
of Waste Water by Addition of Organic Chemicals", Proc. 24th Ind.
Waste Conf., Purdue Univ., 1271-1285, 1969.
64. Tamblyn, T.A. and Sword, B.A., "The Anaerobic Filter for the Denitri-
fication of Agricultural Subsurface Drainage", Proc. 24th Ind. Waste
Conf., Purdue Univ., 1135-1150, 1969.
65. Adams, C.E., Krenkel, P.A., Eckenfelder, W.W., and Bingham, E.G.,
"Removal and Recovery of High Concentrations of Nitrogen", Paper
presented at International Congress on Industrial Waste Water,
Stockholm, Sweden, 1970.
66. Carry, C.F., English, J.N., Massee, A.N., and Pitkin, J.A., "Waste
Water Denitrification in Sand and Carbon Columns", Paper presented
at the 43rd WPCF Conference, Boston, 1970.
287
-------
67. Seidel, D.F. and Crites, R.W.,"Evaluation of Anaerobic Denitrifica-
tion Processes", JSED-ASCE 96, SA2, 267-277, 1970.
68. Wuhrmann, K., "Stickstoff - und Phosphorelimination Ergebnisse von
Versuchen in Technischen Masstab.", Schweiz. Z^. Hydro!. 26_, 520-558,
1964.
69. Slechta, A.F. and Gulp, G.L., "Water Reclamation Studies at the South
Tahoe Public Utility District", JWPCF 39, 787-814, 1967.
70. Balakrishnan, S., "Nitrogen Relationships in Biological Treatment
Process - III. Denitrification in the Modified Activated Sludge
Process", Water Research 3, 177-187, 1969.
71. Johnson, W.K. and Schroepfer, G.J., "Nitrogen Removal-Nitrification
and Denitrification", JWPCF 36, 1015-1036, 1964.
72. Bringmann, G., "Model! Versuche zur Biologischen Stickstoff Abgasung
aus Klarwassern", Gesundheitsingim'eur 81, 140-142, 1959.
73. Bringmann, G., "Optimall Stickstoff - Abgasung Durch Einsatz von
Nitrifikrendem Belebstchlamm und Redox - Steurung", Gesundheit-
singenieur 81, 140-142, 1960.
74. Bringmann, G., "Vollustandige Biologische Stickstoff Eliminierung
aus Klarwasseren im Auschluss an ein Hockeleistuns - Nitrifikations -
Verfahren", Gesundheitsingenieur 82, 233-235, 1961.
75. Earth, E.F., Brenner, R.C. and Lewis, R.F., "Chemical-Biological Con-
trol of Nitrogen and Phosphorus in Waste Water Effluent", JWPCF 40,
2040-2054, 1968.
76. Mulbarger, M.C., "Modifications of the Activated Sludge Process for
Nitrification and Denitrification", Paper presented at the 43rd WPCF
Conference, Boston, 1970.
77- Echelberger, W.F. and Tenney, M.W., "Waste Water Treatment for Com-
plete Nutrient Removal", Wat. Sew. Works, 296-302, 1969.
78. Ludzack, F.J. and Ettinger, M.B., "Controlling Operation to Minimize
Activated Sludge Effluent Nitrogen", JWPCF 34. 920-931, 1962.
79. Okey, R.W., Rickels, R.N. and Taylor, R.B., "Relative Economics of
Animal Waste Disposal by Selected Wet and Dry Techniques in Animal
Waste Management", Cornell Conference on Agricultural Waste Manage-
ment, 369-387, 1969.
80. Symons, J.M., McKinney, R.E., Smith, R.M. and Donovan, E.S., "Degra-
dation of Nitrogen Containing Organic Compounds by Activated Sludge",
Adv. Biol. Waste Treatment, Proc. 3rd Conf. Biol. Waste Treatment,
Ed. W.W. Eckenfelder and J. McCabe, Pergamon Press, 1963.
288
-------
81. Schloesing, T., and Muntz, A., "Sur la Nitrification par les Ferments
Organises", Compt. Rend. Acad. Sci. Paris 84, 301-303, 1873.
82. Warrington, R., "On Nitrification - Part 4", J. Chem. Soc. 59,
484-529, 1891. ~
83. Winogradsky, S., "Recherches sur les Organismes de la Nitrification",
Ann. Inst. Pasteur 4, 213-231, 257-275, 760-771, 1890.
84. Bergey's Manual of Determinative Bacteriology, 7th Ed., Breed, R.S.,
Murray, R.G. and Smith, N.R., The Williams and Wilkins Co., Baltimore,
1957.
85. Waterbury, J.B., Remsen, C.C., and Watson, S.W., "Nitrospira Sp. and
Nitrococcus Sp. Two New Genera of Obligate Autotrophic Nitrite Oxi-
dizing Bacteria", Bact. Proc., 45_, 1968.
86. Verstraete, W.H., "Heterotrophic Nitrification by Arthrobacter Sp.",
Ph.D. Thesis, Cornell Univ., Ithaca, New York, 1971.
87. Marshall, K.C. and Alexander, M., "Nitrification by Aspergillus flavus",
J_. Bacteriol. 82[, 572-575, 1962.
88. Schmidt, E.L., "Cultural Conditions Influencing Nitrate Formation by
Aspergillus flavus", J_. Bact. 79_, 553-558, 1960.
89. Meyerhof, D., "Untersuchengen uber den Atmungsvorgang Nitrification
de Bakterien", Pflug Arch. Des Physio!. 164. 352, 165, 229, 166, 240,
1916.
90. Jensen, H.L., "Effect of Organic Compounds on Nitrosomonas", Nature
165, 974, 1950.
91. Wimmer, G., "Beitrag zur Kenntnis der Nitrifikations Bakterien", Hyg.
Infektionskrankh 48, 135-174, 1904.
92. Stevens, F.L. and Withers, W.A., "Studies in Soil Bacteriology. I.
Nitrification in Soils and Solutions", Zent. Bakt. ]_L 23_, 355-372,
1909.
93. Stevens, F.L. and Withers, W.A., "Studies in Soil Bacteriology. IV.
The Inhibition of Nitrification by Organic Matter Compared in Soils
and Solutions", Zent. Bakt. 27_, 169-186, 1910.
94. Balakrishnan, S. and Eckenfelder, W.W., "Nitrogen Relationships in
Biological Treatment Processes - II. Nitrification in Trickling
Filters", Water Research 3. 167-174, 1969.
95. Balakrishnan, S. and Eckenfelder, W.W., "Nitrogen Relationships in
Biological Treatment Processes - I. Nitrification in the Activated
Sludge Process", Water Research 3, 73-81, 1969.
289
-------
96. Downing, A.L., Painter, H.A., and Knowles, G., "Nitrification in the
Activated Sludge Process", J_. Proc. Inst. Sew. Pur if.. 130-153, 1964.
97. Mattingly, G.E.G., "Studies on Composts Prepared from Waste Materials.
III. Nitrification in Soil", J_. Sci. Food Agriculture 7. 601-605,
1956.
98. Wuhrmann, K., "Objectives: Technology, and Results of Nitrogen and
Phosphorus Removal Processes", in Adv. Hater Quality Improvement,
Ed. E.F. Gloyna, and W.W. Eckenfelder, Univ. of Texas Press, 1968.
99. Loehr, R.C., Anderson, D.F., and Anthonisen, A.C., "An Oxidation
Ditch for the Handling and Treatment of Poultry Wastes", Paper pre-
sented at the Int. Symposium on Livestock Wastes, Columbus, Ohio,
1971.
100. Ruban, E.L., "Nitrogen Metabolism of Nitrosomonas europea", Mikro-
biologia 27, 536-541, 1958.
101. Delwiche, C.C. and Finstein, M.S., "Carbon and Energy Sources for the
Nitrifying Autotrophs, Nitrobacter", J. Bacteriol. 90, 102-107, 1965.
102. Smith, A.J. and Hoare, D.S., "Acetate Assimilation by Nitrobacter
agilis in Relation to its Obligate Autotrophy", J_. Bact. 95_, 844-855,
1968.
103. Bomeke, H., "Uber Die Ernahrungs und Wachstums Faktoren der Nitrifi-
kations", Bakterien Arch Microbiol. 1_4_, 63-70, 1949.
104. Nelson, D.H., "Isolation and Characterization of Nitrosomonas and
Nitrobacter", Zent. Bacteriol: I_, 83_, 280-311, 1931.
105. Winogradsky, S. and Omeliansky, V., "Uber der Influss der Organischen
Substangen aus die Arbeit der Nitrifizierenden Bakterien", Zent.
Bakteriol. Parasitenk. 5_, 329-343, 377-387, 429-440, 1899.
106. Meiklejohn, J., "The Isolation of Nitrosomonas europea in Pure Culture",
J_. Gen. Microbiol. 4_, 185-190, 1950.
107. Omeliansky, V., "Uber die Isolierung der Nitrifikations Microben aus
dem Erdboden", Zent. Bakteriol. Parasitenk, 652-665, 1899.
108. Kingma Boltjes, T.Y., "Untersuchungen uber die Nitrifizierenden
Bakterien", Arch Microbiol. 6_, 79-89, 1935.
109. Cunningham, J.D., "Studies on Microfloral Changes in Developments in
Soil Samples Undergoing Treatment by the Lees and Quastel Percolation
Technique", M.S.A. Thesis, Univ. of Toronto, Canada, 1950.
110. Finstein, M.S., "Terrestrial and Marine Autotrophic Ammonium Oxidizing
Bacteria in Waters of Different Salinities", Bact. Proc., 25, 1969.
290
-------
111. Stojanovic, B.J. and Alexander, M., "Effect of Inorganic Nitrogen on
Nitrification", Soil Sci . Proc. 86_, 208-215, 1958.
112. Alexander, M., "Nitrification" in Soil Nitrogen. Agr. Soc. Agron.,
Publ. 10, Ed. W.V. Bartholomew and F.E. Clark, 307-343, 1965.
113. Montgomery, H.A.G. and Borne, B.J., "The Inhibition of Nitrification
in the BOD Test", Jour, and Proc. of Inst. Sew. Purif., 357-368, 1966.
114. Siddiqui, R.H., Speece, R.E., Engelbrecht, R.S. and Schmidt, J.W.,
"Elimination of Nitrification in the BOD Determination with 0.1M
Ammonia Nitrogen", JWPCF, 39^:4, 579-587, 1967.
115. Young, J.C., "Chemical Methods for Nitrification Control", Proc. 24th
Industrial Waste Conference, Purdue Univ., 1090-1102, 1969.
116. Downing, A.L., Tomilinson, T.G., and Truesdale, G.A., "Effect of
Inhibitors on Nitrification in the Activated Sludge Process", J_.
Proc. Inst. Sew. Purif., 537-554, 1964.
117. Quastel, J.H. and Schoelfield, P.G., "Biochemistry of Nitrification
in Soil", Bact. Rev. 15_, 1-53, 1951.
118. Delwiche, C.C., "Nitrification", A Symposium on Inorganic Nitrogen
Metabolism, Ed. W.D. McElroy and B. Glass, 218-232, 1956.
119. Gibbs, M. and Schiff, J.A., "Chemosynthesis, The Energy Relationships
of Chemoautotrophic Organisms" in Plant Physiology, Academic Press,
New York, Vol. 1, 279-319, 1960.
120. Edwards, J.B. and Robinson, J.B., "Changes in Composition of Contin-
uously Aerated Poultry Manure with Special Reference to Nitrogen",
Proc. Cornell Conference on Agricultural Waste Management, 178-184,
1969.
121. Heukelekian, H., "The Influence of Nitrifying Flora, Oxygen and
Ammonia Supply on the Nitrification of Sewage", Sew. Works J. ]£,
964-979, 1942.
122. Meera Bai, B., Viswanathan, C.V. and Pillai, S.C., "Factors Influencing
Nitrification by Activated Sludge", J_. Sci. Ind. Res. 2J_, 48-55, 1962.
123. Phelps, E.B., Stream Sanitation, John Wiley and Sons, New York, 86, 1960
124. Downing, A.L. and Hopwood, A.P., "Some Observations on the Kinetics
of Nitrifying Activated Sludge Plants", Schweiz. Z_. Hydro!., 26_,
271-280, 1964.
125. Orford, H.E., Heukelekian, H. and Isenberg, E., "Effect of Sludge
Loading and Dissolved Oxygen on the Performance of the Activated Sludge
Process", Adv. Biol. Waste Treatment, Proc. 3rd Conf. Biol. Waste
Treatment, Ed. W.W. Eckenfelder and J. McCabe, Pergamon Press, 1963.
291
-------
126. Wild, H.E., Sawyer, C.N., and McMahon, T.C., "Factors Affecting Nitri-
fication Kinetics", Paper presented at the 43rd WPCF Conference, Boston,
1970.
127. Painter, H.A. and Jones, K., "The Use of the Wide-Bore Dropping Mercury
Electrode for the Determination of Rates of Oxygen Uptake and of Oxi-
dation of Ammonia by Microorganisms", J. Appl. Bact. 26_, 471-483, 1963.
128. Downing, A.L. and Scragg, L.J., "The Effect of Synthetic Detergents
on the Rate of Aeration in Diffused-Air Activated Sludge Plants",
Water Waste Treatment J. 7_, 102-107, 1958.
129. Wuhrmann, K., "Effects of Oxygen Tension on Biochemical Reactions in
Sewage Purification Plants", Adv. in Biol. Waste Treatment. Proc. 3rd
Conf. Biol. Waste Treatment, Ed. W.W. Eckenfelder and J. McCabe,
Pergamon Press, 27-28, 1963.
130. Butt, W.D., and Lees, H., "Oxygen Tension and the Inhibition of
Nitrate Reduction by Nitrobacter", Biochem. J_. 69_, 3-4, 1958.
131. Wezernak, C.T. and Gannon, J.J., "Oxygen Nitrogen Relationships in
Autotrophic Nitrification", Appl. Microbiol. 1_5_, 1211-1215, 1967.
132. Painter, H.A., "A Review of Literature on Inorganic Nitrogen Metabo-
lism in Microorganisms", Water Research 4, 393-450, 1970.
133. Deppe, K. and Engel, H., "Untersuchungen uber die Temparaturabhan-
bigkeit der Nitratbildung durch Nitrobacter winogradski", Zent. Bakt.
Parasitkde II 113, 561-568, 1960.
134. Loveless, J.E. and Painter, H.A., "Influence of Metal Ion Concentration
and pH Value on the Growth of Nitrosomonas Strain Isolated from Acti-
vated Sludge", J. Gen. Microbiol. 52, 1-14, 1968.
135. Buswell, A.M., Mueller, H.F. and Meter, I.V., "Bacteriological Expla-
nation of Rate of Oxygen Consumption in the BOD Test", Sew. Ind.
Waste 26, 276-285, 1954.
136. Laudelot, H. and Van Tichelen, L., "Kinetics of the Nitrite Oxidation
by Nitrobacter winogradski", J_. Bact. 79_, 39-49, 1960.
137. Gibbs, W.M., "The Isolation and Study of Nitrifying Bacteria", Soil
Sci. Proc. 8_, 427-436, 1919.
138. Birch, H.F., "Further Observations on Humus Decomposition and Nitri-
fication", Plant and Soil 11, 262-286, 1959.
139. Agarwal, A.S., Singh, B.R., and Kanehiro, Y., "Soil Nitrogen and
Carbon Mineralization as Effected by Drying-Rewetting Cycles", Soil
Sci. of America Proc. 35_, 95-100, 1971.
292
-------
140. Rimer, A. and Woodward, R.L., "Two Stage Activated Sludge Pilot Plant
Operations - Fitchburg, Massachusetts", Paper presented at 43rd WPCF
Conference, Boston, 1970,
141. Anderson, O.E. and Boswell, F.C., "The Influent of Low Temperature
and Various Concentrations of Ammonium Nitrate on Nitrification in
Acid Soils", Soil Sci . Soc. Amer. Proc. 28_, 525-529, 1964.
142. Anderson, O.E., Boswell, F.C. and Harrison, R.M., "Variations in Low
Temperature Adaptability of Nitrifiers in Acid Soil", Soil Sci. Soc.
Amer. Proc. 35_, 68-71, 1971.
143. Ludzack, F.J., Schaffer, R.B. and Ettinger, M.B., "Temperature and
Feed as Variables in Activated Sludge Performance", JWPCF 33, 141-156,
1961. ~
144. Winogradsky, S. and Winogradsky, H., "Etudes sur la Microbologie du
Sol. 7 Nouvelles Recherches sur les Organismes de la Nitrification",
Ann. Inst. Pasteur 50, 350-432, 1933.
145. Engel, M.S. and Alexander, M., "Autotrophic Oxidation of Ammonium and
Hydroxylamine", Soil Sci. Soc. Agr. Proc. 2A_, 48-50, 1960.
146. Boon, B. and Laudelout, H., "Kinetics of Nitrite Oxidation by Nitro-
bacter Winogradsky", Biochem. J. 85_, 440-447, 1962.
147. Aleem, M.I.H., "The Physiology and Chemoautotrophic Metabolism of
Nitrobacter agilis", Ph.D. Thesis Cornell Univ., 1959.
148. Boulanger, E. and Massol, L., "Etudes sur la Microbes Nitrificateurs",
Ann. Inst. Pasteur, 18, 181-196, 1904.
149. Pokallus, R.S., "Toxicity of Nitrite to Nitrosomonas europea", Ph.D.
Thesis, Rutgers Univ., New Jersey, 1963.
150. Lewis, R.F. and Pramer, D., "loslation of Nitrosomonas in Pure Culture",
J_. Bacteriol. 76_, 524-528, 1959.
151. Reed, L.W. and Sabbe, W.E., "Investigations Concerning a Van Slyke
Reaction Involving Urea and Nitrite", Agron. Absts. 2A_, 1963.
152. Meikeljohn, J., "Aerobic Denitrification", Ann. Appl. Biol. 27_, 558-
573, 1940.
153. Smith, D.H. and Clark, F.C., "Volatile Losses of Nitrogen from Acid
or Neutral Soils or Solutions Containing Nitrite and Ammonia Ions",
Soil Sci. Soc. Agr. Proc. 90_, 86-92, 1960.
154. Verhoeven, W., "Some Remarks on Nitrate and Nitrite Metabolism in
Microorganisms", in A Symposium on Inorganic Nitrogen Metabolism, Ed.
W.D. McElroy and B. Glass, 61-86, John Hopkins Press, Baltimore, 1956.
293
-------
155. Wijler, J. and Delwiche, C.C., "Investigations on the Denitrifying
Process in Soil", Plant and Soil 17, 267-270, 1954.
156. Schwartzbeck, R.A., MacGregor, J.V. and Schmidt, E.L., "Gaseous
Nitrogen Losses from Nitrogen Fertilized Soils Measured with Infra-
red and Mass Spectroscopy", Soil Sci. Soc. Amer. Proc. 25^ 186-195,
1961.
157. Gayon, V. and DuPetit, G., "Recherches sur la Reduction des Nitrates
par les Infiniments Petits", Soc. Sci. Phys. Nat. Bordeaux. Ser. #3_,
i, 201-207, 1886.
158. Johnson, W.K., "Removal of Nitrogen by Biological Treatment", Adv.
in Hater Quality Improvement. Univ. of Texas Press, Ed. E.F. GToyna
and H.W. Eckenfelder, 178-189, 1968.
159. Scheltinga, H.M.J., "Biological Treatment of Animal Wastes", ASAE
Publication SP 0366, Amer. Soc. Agric. Engr., 140-143, 1966.
160. Zehender, F., "Elimination of Nitrogen and Phosphorus in Biological
Sewage Treatment Plants", Proc. Internat. Assoc. Theor. Appl. Limn.
TO., 597-601, 1948.
161. Truesdale, G.A., Wilkinson, R. and Jones, K., "A Comparison of the
Behavior of Various Media in Percolating Filters", Engrs. J_. 60,
273-287, 1961.
162. Barth, E.F., Mulbarger, M., Sal otto, B.V. and Ettinger, M.B., "Removal
of Nitrogen by Municipal Waste Water Treatment Plants", JWPCF 38.
1208-1219, 1966.
163. Bremner, J.M. and Shaw, K., "Denitrification in Soil II. Factors
Affecting Denitrification", J. of Agr. Sci. 51, 40-52, 1958.
164. Woldendrop, J.W., "The Quantitative Influence of the Rhizosphere on
Denitrification", Plant and Soil 17, 267-270, 1962.
165. Meek, B.D., Grass, L.B., Willardson, L.S., and MacKenzie, A.J.,
"Nitrate Transformations in a Column with Controlled Water Table",
Soil Sci. Soc. Amer. Proc. 34_, 235-239, 1970.
166. Mechalas, B.J., Allen, P.M., and Matyskiela, W.W., "A Study of Nitri-
fication and Denitrification", Rept. from Contract 14-12-498, FWQA,
USDI, 1970.
167. Thacker, H.R., "The Reduction of High Nitrate Nitrogen Concentrations
in Natural Waters", Ph.D. Thesis, Virginia Polytech. Inst., 1965.
168. Nilsson, E.S. and Westberg, N., "Bacterial Denitrification of Nitrate
Containing Waste Waters", Vatten (Swedish) 23_, 35-40, 1967.
294
-------
169. Skerman, V.B.D. and MacRae, I.e., "Influence of Oxygen Availability
on the Degree of Nitrate Reduction by Pseudomonas dem'trificans",
Can. J_. Microbiol. 3., 505-530, 1957.
170. Skerman, V.B.D., and MacRae, I.C., "Influence of Oxygen on Reduction
of Nitrate by Adapted Cells of Pseudomonas denitrificans" , Can. J.
Microbiol. 3., 215-230, 1957. - ~
171. Skerman, V.B.D., Carey, R.J. and MacRae, I.e., "Influence of Oxygen
on the Reduction of Nitrate by Washed Suspensions of Adapted Cells
of Achromobacter liquefaciens". Can. J. Microbiol. 4, 243-256, 1958.
172. Kefauver, M. and Allison, F.E., "Nitrite Reduction by Bacterium
denitrificans in Relation to Oxidation-Reduction Potential and
Oxygen Tension", J_. Bact. 73_, 8-14, 1956.
173. McCarty, P.L., "Energetics and Biological Growth", Paper presented
at 5th Rudolf Research Conf., Rutgers Univ., 1969.
174. Koraskova, M.P., "Effect of Aeration on the Process of Bacterial
Nitrate Reduction", Mikrobial 10, 163-178, 1941, Chem. Absts. 36,
4848, 1942.
175. Marshall, R.O., Dishburger, H.J., MacVicar, R., and Hallmark, G.D.,
"Studies on the Effect of Aeration on Nitrate Reduction by Pseudomonas
Species Using N15", 0_. Bact. 6_, 254-258, 1953.
176. Sacks, I.E., and Barker, H.A., "Influence of Oxygen on Nitrate and
Nitrite Reduction", J_. Bact. 58., 11-22, 1949.
177. Wuhrmann, K. and Mechsner, K., "Uber den Einfluss von Sauerstoff-
spannung and Wasserstoffionen Kongentration des Milieus auf die
Mikrobielle Denitrifikation", Pathol . Microbiol. 28_, 99-106, 1965.
178. Valera, C.L. and Alexander, M., "Nutrition and Physiology of Denitri-
fying Bacteria", Plant and Soil 15, 268-280, 1961.
179. Nommik, H., "Investigations on Denitrifi cation in Soil", Acta Agri .
Scand. 2., 195-228, 1956.
180. Khan, M.F.A. and Moore, A.M., "Denitrifying Capacity of Some Alberta
Soils", Can. J_. SpJl ScJ_. , 48_, 89-91, 1968.
181. Jansson, S.L. and Clark, F.E., "Losses of Nitrogen During Decompo-
sition of Plant Material in the Presence of Inorganic Nitrogen",
Sc. Soc. Amer. Proc. 16., 330-334, 1952.
182. Bollag, J.M., Orcutt, M.L. and Bollag, B., "Denitrification by Iso-
lated Soil Bacteria Under Various Environmental Conditions", Soil
Sci. Soc. Amer. Proc. 34, 875-879, 1970.
295
-------
183. Pearsall, W.H. and Mortimer, C.H., "Oxidation Reduction Potentials
in Water Logged Soils, Natural Waters, and Muds", J_. Ecol. 27_,
483-501, 1939.
184. Patrick, W.H., "Nitrate Reduction Rates in Submerged Soil as Affected
by Redox Potential", Trans. 7th_ Ijvt. Congress Soil Sci. 2_, 494-500,
1960.
185. Zuckerman, M.M. and Molof, A.H., "High Quality Reuse Water by Chemical
Physical Wastewater Treatment", JHPCF 42, 437-456, 1970.
186. Taylor, R.L., Mathews, E.R. and Christenson, C.N., "Effects of High
Concentrations of Nitrogen on Activated Sludge", _Sj3w_.anc[ IjicL Wastes
28, 177-182, 1956.
187. Wuhrmann, K., "Nitrogen Removal in Sewage Treatment Processes", Verh.
Int. Verein Limnol., Ij^, 580-596, 1964.
188. Jensen, H.L., "The Microbiology of Barnyard Manure. I. Changes in
the Microflora and Their Relation to Nitrification", J_. Agr. Sci. 21_,
38-80, 1931.
189. Jensen, H.L., "The Microbiology of Farmyard Manure Decomposition in
Soil. III. Decomposition of the Cells of Microorganisms", J_. Agr.
Sci. 22_, 1-25, 1932.
190. Bremner, J.M. and K. Shaw, "The Mineralization of Some Nitrogenous
Materials in Soil", J_. Sci. Food Agr. 8_, 341-347, 1957.
191. McKenzie, H.A. and Wallace, H.S., "The Kjeldahl Determination of
Nitrogen: A Critical Study of Digestion Conditions - Temperature,
Catalyst, and Oxidizing Agent", Aust. J_. Chem. 7_, 55-71, 1954.
192. Montgomery, H.A.C. and Dymock, J.F., "The Determination of Nitrite
in Water", Analyst 86, 414-416, 1961.
193. Methods of Soil Analysis, Ed. Black, C.A., Evans, D.D., White, J.L.,
Ensminger, I.E. and Clark, F.E., Amer. Soc. Agron. Inc., Madison,
Wise.,1965.
194. Papanos, S. and Brown, B.A., "Poultry Manure", Connecticut Aqric.
Exp. Stat. Bui. 272, 1950.
296
-------
APPENDIX
TABLE PAGE
Effectiveness of Contaminant Removal Results of 294
Jar Tests Phosphate Removal Study
II Values of F at Different pH and Temperatures 312
III Values of F and Corresponding Values of (1-F)/F 313
IV Values of L at Different Temperatures and pH Values 314
V Example Calculations to Obtain K~ Values Using 315
Experimental Data
VI Batch Study Nitrification - Operational Data - 319
Units b, c, d, e, 40, 50, 60, 70
VII Observed and Theoretical Values of COD for the 331
Observed N02-N and N03-N Decrease at 20°C and 35°C
VIII Computations of the Amount of Glucose and Chicken 333
Manure Based on Their Theoretical COD -
Denitrification Run VII
Evaluation of Analytical Methods to Determine NH^-N, 335
N02-N, and N03~N in Poultry Manure
297
-------
ro
10
oo
TABLE I
EFFECTIVENESS OF CONTAMINANT REMOVAL -
RESULTS OF JAR TESTS - PHOSPHATE REMOVAL STUDY
RUN JAR
1
1
2
3
4
5
6
2
1
2
3
4
5
6
3
1
2
3
4
5
6
CHEMICAL
USED FINAL
(mg/1) pH
alum - poultry
50
100
150
200
250
300
alum - poultry
100
120
140
160
180
200
alum - poultry
200
240
280
320
360
400
manure
7.2
7.1
7.5
7.3
7.2
7.0
manure
7.1
7.2
7.0
6.9
6.8
6.7
manure
6.9
6.8
6.6
6.6
6.5
6.5
ORTHO-PHOSPHATE
removed
(%)
solution
39
66
73
83
90
91
solution
26
45
57
62
66
73
solution
71
76
79
82
85
92
remaining
(mg/1)
22.5
12.5
10.0
6.5
3.8
3.5
30.0
22.5
17.5
15.5
14.0
11.0
21.3
17.5
15.0
13.5
11.0
6.0
TOTAL PHOSPHATE
removed
(%)
27
62
79
_
-
-
29
46
62
62
64
77
39
45
59
75
78
85
remaining
(mg/1)
35.0
18.0
10.0
-
-
-
48.8
37.5
26.3
26.0
25.0
16.0
53.8
48.8
36.3
22.0
19.0
13.5
-------
TABLE I continued.
RUN
4
5
6
7
JAR
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
CHEMICAL
USED
(mg/1)
alum - poultry
300
340
380
420
460
500
alum - poultry
600
680
760
840
920
1000
alum - poultry
1200
1360
1520
1680
1840
2000
alum used, pH
100
120
140
160
180
200
ORTHO-PHOSPHATE
FINAL
PH
manure
7.1
7.1
7.0
6.6
6.5
6.7
manure
6.6
6.5
6.4
6.3
6.3
6.2
manure
6.5
6.4
6.3
6.2
6.1
6.0
adjusted
6.5
6.4
6.3
6.3
6.2
6.2
removed
(%)
solution
65
79
80
85
89
93
solution
77
79
83
89
92
95
solution
87
91
92
94
95
97
to 7.0 before
43
54
70
79
88
91
TOTAL PHOSPHATE
remaining removed remaining
(mg/1) («) (mg/1)
21.0
12.5
12.0
9.3
6.8
4.5
38.8
33.8
27.5
18.0
13.5
9.0
55.0
37.5
35.0
25.0
19.0
12.5
alum added -
23.0
18.8
12.0
8.5
4.8
3.8
67
76
82
80
91
93
76
80
86
90
94
95
83
89
90
92
95
96
poultry manure solution
42
50
69
78
84
90
29.0
21.5
16.0
17.8
7.8
6.0
67.5
57.5
40.0
27.0
18.0
14.5
90.0
60.0
55.0
41.3
26.3
22.5
35.0
30.0
18.8
13.0
9.5
6.0
-------
TABLE I continued.
GO
o
o
RUN
8
9
9a
9b
JAR
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
CHEMICAL
USED
(mg/1 )
alum used,
100
120
140
160
180
200
alum used,
100
120
140
160
180
200
alum used,
100
120
140
160
180
200
alum used,
100
120
140
160
180
200
ORTHO-PHOSPHATE
FINAL
PH
pH adjusted
5.7
5.6
5.6
5.5
5.4
5.4
pH adjusted
5.8
5.8
5.8
5.7
5.7
5.6
pH adjusted
5.5
5.4
5.4
5.3
5.3
5.2
pH adjusted
-
-
-
-
-
-
removed
(%)
to 6.5 before
57
74
84
92
94
95
to 6.0 before
60
61
85
91
93
95
to 6.0 before
59
75
92
93
96
96
to 6.0 before
63
74
90
90
98
99
TOTAL PHOSPHATE
remaining removed remaining
(mg/1) (%) (mg/1)
alum added -
20.0
12.0
7.5
3.7
3.0
2.5
alum added -
21.0
20.0
8.0
4.8
3.5
2.8
alum added -
19.5
12.0
4.0
3.3
2.0
1.9
alum added -
17.5
12.2
4.8
4.9
1.0
0.5
poultry manure solution
62
75
85
90
93
93
poultry manure solution
55
58
82
88
90
92
poultry manure solution
60
64
81
88
92
93
poultry manure solution
-
_
_
_
_
_
24.0
16.0
9.5
6.6
4.6
4.3
34.0
32.0
14.0
9.3
7.8
5.8
26.0
23.0
12.0
7.5
5.0
4.8
_
_
_
«.
_
_
-------
TABLE I continued.
CO
o
RUN
10
lOa
11
12
JAR
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
CHEMICAL
USED
(mg/1)
alum used,
100
120
140
160
180
200
alum used,
100
120
140
160
180
200
alum used,
300
340
380
420
460
500
alum used,
300
340
380
420
460
500
ORTHO-PHOSPHATE
FINAL
PH
pH adjusted
5.3
5.3
5.2
5.0
5.1
5.1
pH adjusted
-
-
-
-
_
-
pH adjusted
6.5
6.4
6.3
6.2
6.1
6.0
pH adjusted
6.1
6.1
5.9
5.8
5.8
5.8
removed
(50
to 5.5 before
61
83
89
94
95
-
to 5.5 before
72
87
96
97
98
98
to 7.0 before
74
79
86
90
94
95
to 6.5 before
86
93
95
96
97
98
TOTAL PHOSPHATE
remaining removed remaining
(mg/1) (%) (mg/1)
alum added -
18.5
8.0
5.5
2.8
2.5
-
alum added -
13.0
6.0
2.0
1.5
0.8
0.8
alum added -
18.5
14.5
10.0
7.0
4.5
3.5
alum added -
11.0
6.0
4.0
3.5
2.3
2.0
poultry manure solution
57
81
85
92
91
93
poultry manure solution
-
-
-
-
_
-
poultry manure solution
75
83
86
92
94
95
poultry manure solution
84
91
94
95
96
97
29.5
13.0
10.5
5.8
6.0
5.0
-
-
-
-
-
-
25.0
17.5
14.5
5.0
6.5
5.0
15.5
9.0
5.5
4.5
3.5
3.3
-------
TABLE I continued.
GO
o
ro
RUN JAR
13
1
2
3
4
5
6
14
1
2
3
4
5
6
15
1
2
3
4
5
6
16
1
2
3
4
5
6
CHEMICAL
USED
(mg/1)
alum used,
300
340
380
420
460
500
alum used,
300
340
380
420
460
500
alum used,
600
680
760
840
920
1000
alum used,
600
680
760
840
920
1000
ORTHO-PHOSPHATE
FINAL
PH
pH adjusted
5.9
6.0
5.9
5.8
5.6
5.6
pH adjusted
5.5
5.4
5.3
5.3
5.2
5.0
pH adjusted
6.1
6.0
5.9
5.8
5.7
5.5
pH adjusted
5.9
5.8
5.8
5.7
5.5
5.5
removed
(%)
to 6.0 before
89
91
94
96
96
97
to 5.5 before
94
97
97
97
97
97
to 7.0 before
67
76
82
87
89
97
to 6.5 before
75
83
86
93
94
96
TOTAL PHOSPHATE
remaining removed remaining
(mg/1) (%) (mg/1)
alum added -
9.0
7.5
5.0
3.5
3.4
2.6
alum added -
5.0
2.9
2.4
2.5
2.3
2.7
alum added -
42.5
30.0
23.7
16.5
14.5
4.0
alum added -
33.8
22.5
18.8
9.0
7.5
6.0
poultry manure solution
87
90
94
94
97
97
poultry manure solution
93
95
96
96
97
98
poultry manure solution
75
83
84
90
91
94
poultry manure solution
79
85
86
93
92
95
13.5
9.8
6.0
6.5
3.0
2.8
6.2
4.6
3.8
3.8
2.4
1.5
43.8
30.0
28.8
18.0
16.5
10.5
37.5
27.5
25.0
12.0
14.5
10.0
-------
TABLE I continued.
CO
o
CO
RUN
17
18
19
20
JAR
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
CHEMICAL
USED
(mg/1)
alum used,
360
440
520
600
680
760
alum used,
360
440
520
600
680
760
alum used,
720
880
1040
1200
1360
1520
alum used,
720
880
1040
1200
1360
1520
ORTHO-PHOSPHATE
FINAL
pH
pH adjusted
-
-
-
-
-
-
pH adjusted
5.3
5.2
5.2
4.9
4.8
4.7
pH adjusted
6.6
6.5
6.3
6.4
6.2
6.1
pH adjusted
-
-
-
-
-
-
removed
(%)
to 6.0 before
62
75
81
89
92
95
to 5.5 before
77
88
91
94
93
87
to 7.0 before
62
72
81
82
82
87
to 6.5 before
63
71
79
79
85
87
TOTAL PHOSPHATE
remaining removed
(mg/1) (*)
alum added -
45.0
30.0
22.5
13.8
10.0
6.3
alum added -
36.3
18.8
13.8
10.5
11.5
20.0
alum added -
120.0
87.5
60.0
57.5
57.5
42.5
alum added -
115.0
90.0
67.5
66.3
46.3
40.0
poultry manure
64
77
84
84
93
95
poultry manure
70
86
89
92
87
83
poultry manure
66
72
82
83
85
87
poultry manure
63
73
80
78
84
87
remaining
(mg/1)
solution
67.5
42.5
30.0
30.0
12.5
8.8
solution
51.3
23.8
18.8
14.0
21.5
28.5
solution
147.5
120.0
77.5
73.8
66.3
57.5
solution
135.0
100.0
75.0
80.0
57.5
47.0
-------
TABLE I continued.
GO
o
RUN
21
22
23
24
JAR
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
CHEMICAL
USED
(mg/1)
alum used,
720
880
1040
1200
1360
1520
alum used,
720
880
1040
1200
1360
1520
alum used,
720
880
1040
1200
400
600
alum used,
30
60
90
120
150
180
FINAL
PH
pH adjusted
-
-
-
_
-
-
pH adjusted
-
-
-
-
-
-
pH adjusted
-
-
-
-
-
-
pH adjusted
-
-
-
-
-
—
ORTHO-PHOSPHATE
removed
(*)
to 6.0 before
69
68
85
91
92
94
to 5.5 before
59
61
63
78
78
82
to 5.5 before
84
90
90
91
56
79
to 6.5 before
22
22
36
36
46
56
remaining
(mg/D
alum added -
100.0
102.5
47.5
30.0
25.0
18.8
alum added -
300.0
287.5
275.0
160.0
160.0
135.0
alum added -
57.5
37.0
36.0
32.5
160.0
75.0
alum added -
70.0
70.5
57.5
57.5
48.8
39.0
TOTAL PHOSPHATE
removed
(%)
poultry manure
68
65
81
87
91
92
poultry manure
55
67
68
80
82
84
poultry manure
82
89
91
93
52
75
poultry manure
-17
6
9
48
48
52
remaining
(mg/1)
solution
125.0
137.5
72.5
52.5
35.0
31.3
solution
462.5
337.5
325.0
200.0
185.0
160.0
solution
72.5
46.0
37.5
30.0
192.0
100.0
solution
131.0
105.0
102.5
58.8
58.8
53.8
-------
TABLE I continued.
o
tn
RUN
25
26
27
28
JAR
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
CHEMICAL
USED
(mg/1)
alum used,
60
120
180
240
300
360
alum used,
50
100
150
200
250
300
alum used,
200
300
400
500
600
700
alum used,
200
300
400
500
600
700
ORTHO-PHOSPHATE
FINAL
PH
pH adjusted
-
-
-
-
-
-
pH adjusted
-
-
-
-
-
-
pH adjusted
-
-
-
-
-
-
pH adjusted
-
-
-
-
-
-
removed
(%)
to 6.5 before
17
42
57
81
94
97
to 5.5 before
11
25
66
93
93
97
to 6.5 before
34
47
69
83
95
97
to 6.0 before
28
56
75
91
97
98
TOTAL PHOSPHATE
remaining removed
(mg/D (%)
alum added - poultry
75.0
52.5
38.8
17.5
5.2
2.5
alum added - poultry
62.5
52.5
23.7
5.0
4.8
2.5
alum added - poultry
115.0
92.5
55.0
30.0
9.0
5.3
alum added - poultry
135.0
82.5
47.0
16.3
5.5
3.3
manure
7
37
59
75
94
96
manure
13
20
65
87
93
95
manure
59
72
77
88
95
97
manure
39
61
75
90
97
97
remaining
(mg/D
solution
100.0
67.5
43.8
26.3
6.5
4.5
solution
65.0
60.0
26.3
10.0
5.0
3.8
solution
105.0
72.5
60.0
31.3
13.8
8.8
solution
142.5
90.0
57.5
23.8
7.0
6.0
-------
TABLE I continued.
GO
o
en
RUN
29
30
31
32
JAR
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
CHEMICAL
USED
(mg/1 )
alum used, pH
100
200
300
400
500
600
alum used, pH
600
800
1000
1200
1400
1600
alum used, pH
200
400
600
800
1000
1200
alum - poultry
800
1000
1200
1400
1600
1800
ORTHO-PHOSPHATE
FINAL
PH
adjusted
-
-
-
_
_
-
adjusted
-
-
-
-
-
-
adjusted
-
-
-
-
-
-
removed
(%)
to 5.5 before
6
30
60
76
55
68
to 6.5 before
61
70
83
93
95
98
to 5.5 before
21
26
66
82
92
95
TOTAL PHOSPHATE
remaining removed remaining
(mg/1) (%) (mg/1)
alum added -
175.0
130.0
75.0
45.0
8.5
6.0
alum added -
145.0
110.0
62.5
26.2
18.8
10.0
alum added -
290.0
270.0
125.0
65.0
30.0
20.0
poultry manure solution
7
36
60
71
94
96
poultry manure solution
56
71
83
92
93
96
poultry manure solution
21
33
69
84
92
94
195.0
135.0
85.0
60.0
12.0
7.8
165.0
110.0
62.5
31.3
27.5
14.5
340.0
290.0
135.0
70.0
35.0
25.0
manure solution
6.6
6.4
6.2
6.0
5.8
5.7
69
75
89
96
97
99
96.0
75.0
35.0
12.8
9.0
4.0
70
79
89
94
96
98
115.0
80.0
42.5
23.8
15.0
7.0
-------
TABLE I continued.
GO
O
RUN JAR
33
1
2
3
4
5
6
34
1
2
3
4
5
6
35
1
2
3
4
5
6
36
1
2
3
4
5
6
CHEMICAL
USED FINAL
(mg/1) pH
alum -
500
650
700
850
1000
1150
lime -
118
248
395
475
810
1037
lime -
315
735
1500
2096
2000
2260
lime -
605
935
2150
4155
4320
4390
poultry manure
-
_
-
-
-
-
poultry manure
8.4
8.9
9.3
9.8
10.4
10.7
poultry manure
8.5
9.0
9.5
10.0
10.5
11.0
poultry manure
8.4
8.7
9.2
9.7
10.0
10.6
ORTHO-PHOSPHATE
removed
(X)
solution
84
91
94
97
98
98
solution
12
22
52
53
58
67
solution
40
63
53
75
75
82
solution
47
56
61
69
72
80
remaining
(mg/1)
29.5
16.3
10.5
5.0
3.8
4.0
70.0
62.5
38.0
37.5
33.8
26.3
102.0
62.5
45.0
42.5
42.5
31.3
175.0
145.0
130.0
105.0
95.0
67.5
TOTAL
removed
86
93
95
97
98
98
10
29
43
49
62
60
40
62
71
74
75
80
31
49
57
64
65
73
PHOSPHATE
remaining
(mg/1)
35.0
18.8
13.0
8.0
5.8
5.9
95.0
75.0
60.0
53.8
40.0
42.5
115.0
72.5
55.0
48.8
47.5
37.5
255.0
190.0
160.0
135.0
130.0
100.0
-------
TABLE I continued.
co
o
00
CHEMICAL
USED
RUN JAR (mg/1 )
ORTHO-PHOSPHATE
FINAL
PH
removed
(%)
37 lime - poultry manure solution
1 1250 8.5 61
2 3290 9.0 63
3 6350 9.5 67
4 8000 9.9 74
5
6
remaining
(mg/1)
248.0
236.0
210.0
165.0
TOTAL
removed
(%)
-
PHOSPHATE
remaining
(mg/1)
-
38 lime - poultry manure solution
1 1115
2 3177
3 5000
4 5650
5 7000
6 7085
8.4
8.8
9.4
9.7
10.2
10.4
47
64
73
76
83
83
350.0
240.0
180.0
155.0
115.0
110.0
49
70
76
81
85
86
420.0
244.0
200.0
160.0
122.0
115.0
-------
TABLE I continued.
CO
o
RUN JAR
39
1
2
3
4
5
6
40
1
2
3
4
5
6
41
1
2
3
4
5
6
CHEMICAL
USED
(mg/1)
r +++
Fe
50
100
150
200
250
300
0
100
150
200
250
300
0
100
150
200
250
300
lime -
320
380
370
360
380
380
640
500
600
490
500
480
1310
1200
1200
1280
1270
1240
ORTHO-PHOSPHATE
FINAL
PH
lime and
chloride
8.7
8.6
8.6
8.6
8.7
8.6
lime and
chloride
8.9
8.8
8.8
8.8
8.8
8.8
lime and
chloride
9.4
11.0
9.2
9.3
9.3
9.2
removed
(«)
ferric chloride
added - poultry
5
7
18
34
39
44
ferric chloride
added - poultry
53
44
58
62
67
69
ferric chloride
added - poultry
53
68
71
74
79
81
remaining
(mg/1)
used, pH adjusted
manure solution
107.5
105.0
92.5
75.0
68.8
63.8
used, pH adjusted
manure solution
85.0
100.0
75.0
67.5
60.0
55.0
used, pH adjusted
manure solution
80.0
55.0
50.0
45.0
35.0
32.5
TOTAL PHOSPHATE
removed
(%)
to 8.5 before
11
14
24
39
45
47
to 9.0 before
55
49
57
64
68
69
to 9.5 before
56
64
69
72
75
78
remaining
(mg/1)
ferric
150.0
145.0
127.5
102.5
92.5
90.0
ferric
115.0
130.0
110.0
92.5
82.5
80.0
ferric
85.0
70.0
60.0
55.0
50.0
42.5
-------
TABLE I continued.
RUN
JAR
CHEMICAL
USED
(mg/1)
ORTHO-PHOSPHATE
TOTAL PHOSPHATE
FINAL
PH
removed
(*)
remaining
(mg/1)
removed
remaining
(mg/1)
42
43
laboratory accident - no data obtained
alum - poultry manure solution
44
45
46
1
2
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
600
700
alum -
100
150
200
300
400
500
alum -
100
200
300
500
700
900
lime -
70
138
285
630
1214
1606
-
-
poultry manure
-
-
-
-
-
-
poultry manure
-
-
-
-
-
-
67
80
solution
20
33
52
64
75
86
solution
22
31
49
61
72
83
110.3 72
66.0 83
150.0
125.0
90.0
67.5
46.3
27.5
260.0
230.0
170.0
130.0
94.0
57.5
147.3
87.0
-
_
_
_
_
-
_
_
_
_
_
-
dairy manure solution
7.5
8.0
8.5
9.0
9.5
10.0
22
39
64
81
84
84
132.5
102.5
60.0
31.3
27.5
26.3
-
_
_
_
_
-
-------
TABLE I continued.
RUN JAR
47
1
2
3
4
5
6
48
1
2
3
4
5
6
49
1
2
3
4
5
6
50
1
2
3
4
5
6
CHEMICAL
USED FINAL
(mg/1) pH
lime -
138
270
634
1210
2300
2978
alum -
50
100
125
150
200
250
alum -
100
200
250
300
400
500
alum -
200
400
500
600
800
1000
dairy manure
dairy manure
dairy manure
7.0
6.8
6.7
6.6
6.4
6.0
dairy manure
7.0
6.8
6.7
6.6
6.4
5.6
ORTHO-PHOSPHATE
removed
solution
24
51
73
86
86
87
solution
31
60
69
84
95
97
solution
42
73
87
93
98
99
solution
58
75
81
90
93
99
remaining
(mg/1)
250.0
160.0
90.0
48.0
45.0
44.0
39.0
22.5
17.5
90.0
26.0
1.6
65.0
30.0
15.0
8.0
2.0
1.2
100.0
60.0
45.0
25.0
17.5
3.5
TOTAL
removed
44
63
70
83
94
98
50
73
84
87
97
98
38
54
74
84
95
99
PHOSPHATE
remaining
(mg/1)
49.0
32.5
26.3
14.5
5.0
2.0
85.0
45.0
27.5
22.5
6.0
3.4
230.0
170.0
95.0
58.0
18.5
3.6
-------
TABLE I continued.
co
ro
RUN
51
52
53
54
JAR
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
1
2
3
4
5
6
CHEMICAL
USED FINAL
(mg/1 ) pH
lime -
29
54
80
151
218
280
lime -
75
119
182
291
475
800
lime -
105
330
750
1165
lime -
200
455
750
910
1050
1090
dairy manure
-
_
_
_
_
-
dairy manure
-
-
-
-
-
-
dairy manure
-
-
-
-
ORTHO-PHOSPHATE
removed
(%)
solution
17
33
45
47
47
49
solution
40
55
57
59
62
80
solution
34
56
60
67
remaining
(mg/1)
47.0
38.0
31.0
30.5
30.0
28.8
74.0
55.0
52.5
50.0
47.0
25.0
150.0
100.0
90.0
75.0
TOTAL
removed
(%)
23
35
40
50
50
56
36
51
54
59
63
74
36
55
56
65
PHOSPHATE
remaining
(mg/1)
67.0
57.5
52.5
43.8
43.8
38.5
115.0
88.0
83.0
74.0
67.0
47.0
210.0
150.0
145.0
115.0
poultry manure solution
8.2
8.5
9.0
9.4
9.8
9.8
66
85
84
84
92
93
40
17
18
18
9
8.5
71
84
81
83
91
92
47
26
31
27
15
13
-------
TABLE I continued.
OJ
GO
RUN JAR
55
1
2
3
4
5
6
56
1
2
3
4
5
6
CHEMICAL
USED
(mg/1 )
ORTHO-PHOSPHATE
FINAL
PH
removed remaining
W (mg/D
TOTAL
removed
(%)
PHOSPHATE
remaining
(mg/1)
alum - poultry manure solution
100
200
300
400
500
600
non salts
0
50
TOO
150
200
300
6.9
6.8
6.6
6.4
6.2
6.0
r +++
- Fe
7.1
7.5
7.3
7.1
6.8
5.9
29
42
57
67
84
94
poultry manure solution
-
27
36
68
91
98
82
68
50
38
18
7
75
55
48
24
7.4
1.2
12
16
30
35
50
58
-
-
-
-
-
-
125
119
100
92
70
60
-
-
-
-
-
-
-------
TABLE I continued.
RUN JAR
CHEMICAL
USED
(mg/1)
ORTHO-PHOSPHATE TOTAL
FINAL removed
pH (%)
remaining removed
(mg/1) (%)
PHOSPHATE
remaining
(mg/1)
57
co
58
59
1
2
3
4
5
6
1
2
3
4
5
6
Fe
0
100
150
200
250
300
0
100
150
200
250
300
lime - poultry manure solution
Fe addition
- lime to an initial pH of 8.5 followed by
90
90
90
90
90
90
320
320
320
320
320
320
8.0
7.7
7.4
6.9
6.1
5.7
34
42
65
85
97
98
poultry manure solution
iii
Fe addition
8.6
8.2
7.5
7.5
7.2
6.9
71
93
96
98
99
49
43
26
11
2
1.2
lime to an initial
25
6
3
2
.5
-
66
70
89
91
98
99
pH of 9.0
79
94
97
97
99
99
52
47
33
12
3
1.5
followed by
27
8
4
4
1.2
1.5
poultry manure solution - lime to an initial pH of 9.5 followed by
Fe addition
1
2
3
4
5
6
0
50
100
150
200
250
750
750
700
750
730
750
9.3
9.3
8.9
8.6
8.4
8.1
72
90
92
95
99
99
16
5.5
4.7
2.8
.5
.5
87
93
95
97
99
99
17
8.5
6.5
3.3
1.2
1.0
-------
TABLE I concluded.
CO
en
RUN JAR
60
1
2
3
4
5
61
1
2
3
4
5
CHEMICAL
USED
(mg/1)
Fe+++
0
50
100
150
200
0
100
150
200
250
lime -
875
820
830
880
850
150
160
160
165
165
ORTHO-PHOSPHATE
FINAL removed
pH («)
poultry manure
Fe addition
9.8
9.4
9.1
8.9
8.8
poultry manure
Fe addition
8.4
7.5
7.0
6.8
6.6
solution
46
82
92
97
99
solution
44
84
94
95
98
remaining
(mg/1)
- lime to initial
29
9.8
4.5
1.7
.7
- lime to initial
45
13
5
4
1
TOTAL PHOSPHATE
removed
pH of 10.
45
76
87
93
97
pH of 8.5
15
81
91
89
98
remaining
(mg/1)
0 followed by
32
14
7
3
1
f ol 1 owed by
77
17
8
10
2
.5
.7
.8
-------
co
TABLE II
VALUES OF F AT DIFFERENT pH AND TEMPERATURES
TEMPERATURE
10
15
20
25
30
35
7.0
0.002
0.003
0.004
0.005
0.008
0.014
7.5
0.006
0.009
0.012
0.017
0.025
0.043
8.0
0.020
0.028
0.037
0.052
0.076
0.125
8.5
0.061
0.082
0.110
0.148
0.207
0.312
pH
9.0
0.170
0.221
0.280
0.354
0.452
0.589
9.5
0.393
0.473
0.552
0.634
0.723
0.819
10.0
0.672
0.739
0.796
0.846
0.892
0.935
10.5
0.866
0.900
0.925
0.945
0.963
0.978
11.0
0.953
0.966
0.975
0.982
0.988
0.993
-------
TABLE III
VALUES OF F AND CORRESPONDING VALUES OF (1-F)/F
0-F)/F F_ (1
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.10
0.12
0.14
0.16
0.18
0.20
0.22
0.24
0.25
0.26
0.28
99.0000
49.0000
32.3333
24.0000
19.0000
15.6667
13.2857
11.5000
10.1111
9.0000
7.3333
6.1429
5.2500
4.5556
4.0000
3.5455
3.1667
3.0000
2.8462
2.5714
0.30
0.32
0.34
0.36
0.38
0.40
0.42
0.44
0.46
0.48
0.50
0.52
0.54
0.56
0.58
0.60
0.62
0.64
0.66
0.68
2.3333
2.1250
1.9412
1.7778
1.6316
1.5000
1.3810
1.2727
1.1739
1.0833
1.0000
0.9231
0.8519
0.7857
0.7241
0.6667
0.6129
0.5625
0.5152
0.4706
0.70
0.72
0.74
0.75
0.76
0.78
0.80
0.82
0.84
0.86
0.88
0.90
0.92
0.94
0.96
0.98
0.99
1.00
0.4286
0.3889
0.3514
0.3333
0.3158
0.2821
0.2500
0.2195
0.1905
0.1628
0.1364
0.1111
0.0869
0.0638
0.0416
0.0204
0.0101
0
317
-------
TABLE IV
VALUES OF L AT DIFFERENT TEMPERATURES AND pH VALUES
pH TEMPERATURE (°C)
8.0
8.2
8.4
8.6
8.8
9.0
w 9.2
» 9.4
9.6
9.8
10.0
10.2
10.4
10.6
10.8
11.0
11.2
11.4
10
22.3295
22.3412
22.3594
22.3876
22.4307
22.4955
22.5903
22.7242
22.9054
23.1386
23.4236
23.7550
24.1246
24.5231
24.9425
25.3761
25.8193
26.2687
15
22.0122
22.0282
22.0530
22.0912
22.1488
22.1873
22.3552
22.5218
22.7395
23.0096
23.3281
23.6874
24.0783
24.4923
24.9223
25.3631
25.8110
26.2634
18
21.8287
21 .8479
21 .8776
21.9228
21.9907
22.0895
22.2286
22.4157
22.6550
22.9457
23.2821
23.6555
24.0569
24.4782
24.9132
25.3572
25.8072
26.2610
21
21
21
21
21
21
22
22
22
22
23
23
24
24
24
25
25
26
20
.7048
.7265
.7599
.8107
.8862
.9954
.1472
.3487
.6027
.9069
.2545
.6366
.0442
.4699
.9078
.3538
.8050
.2596
21
21
21
21
21
21
22
22
22
22
23
23
24
24
24
25
25
26
22
.5777
.6023
.6400
.6971
.7814
.9019
.0674
.2839
.5528
.8703
.2288
.6191
.0327
.4624
.9030
.3507
.8030
.2584
24
21 .4455
21.4735
21.5163
21.5807
21.6749
21.8081
21 .9885
22.2209
22.5050
22.8357
23.2047
23.6029
24.0220
24.4554
24.8985
25.3478
25.8012
26.2572
26
21.3066
21.3387
21.3876
21 .4605
21.5661
21.7136
21.9102
22.1593
22.4590
22.8029
23.1822
23.5878
24.0121
24.4490
24.8944
25.3452
25.7996
26.2562
30
20.9962
21.0397
21.1051
21.2007
21.3356
21.5180
21.7524
22.0385
22.3709
22.7413
23.1404
23.5602
23.9941
24.4374
24.8870
25.3405
25.7966
26.2543
32
20.8169
20.8688
20.9459
21.0571
21.2113
21.4155
21.6722
21 .9789
22.3285
22.7122
23.1210
23.5474
23.9859
24.4322
24.8836
25.3384
25.7952
26.2534
35
20.4986
20.5692
20.6719
20.8157
21 .0082
21.2532
21.5490
21 .8897
22.2665
22.6704
23.0934
23.5295
23.9743
24.4248
24.8789
25.3354
25.7934
26.2522
-------
TABLE V
EXAMPLE CALCULATIONS TO OBTAIN
KD VALUES USING EXPERIMENTAL DATA
This example describes the approach and methods that were used to
obtain the KQJ temperature, and air flow relationships in a batch experi-
ment in which the pH was uncontrolled.
A sample of poultry manure suspension (500 ml) was aerated after ad-
justing the pH to 10.6 with a solution of sodium hydroxide. The reading
in the air flow meter was 10 SCFH. The rate of aeration was 20 SCFH per
liter of liquid. The total solids content, COD, orthophosphate and ammonia
nitrogen contents, and viscosity of the poultry manure suspension used are
given in Table A.
TABLE A
CHARACTERISTICS OF THE UNAERATED SUSPENSION
OF POULTRY MANURE
Total solids (mg/1) 12900
Chemical Oxygen Demand (mg/1) 415
Orthophosphate (mg P04/1) 70
Ammonia nitrogen (mg N/l) 752
Viscosity (cp) 3
The concentration of ammonia, pH value, and the temperature of the
samples drawn from the ammonia desorption apparatus at different time
intervals are noted in the following Table. The sampling would continue
until sufficient data had been collected to characterize the removal
characteristics in the run. Generally a period of 8 hours was sufficient
although in the pilot plant experiments the sampling continued over
several days.
319
-------
TABLE B
pH VALUE, TEMPERATURE AND CONCENTRATION OF AMMONIA
OF THE LIQUID AFTER DIFFERENT PERIODS OF AERATION
Time of
sampling
(hr)
0
1
2
3
p!
10.6
10.4
10.2
10.1
Temperature
°C
21.5
17.5
16
16
Ammonia
concentration in
(mgN/1)
752
554
436
351
the unit
The value of KQ is related to the concentration of ammonia, pH value,
duration of aeration, and temperature ( Equation 43 ). For the convenience
of calculations, the data in Table B have been rearranged and included in
Table C.
The duration of aeration, in all the three sets of data, was 60 minutes.
The values of L at the different temperatures (i.e., the mean values of the
temperatures) and pH values were obtained by using the tables similar to
Table IV, Appendix.
The observed values for KQ during the three periods of aeration are
included in Table D. These values were obtained by using the previously
acquired data with Equation 43. For comparative purposes, the KD values
predicted by Equation 55 for the data collected in this experiment also
are shown in Table D.
TABLE D
VALUES OF COEFFICIENT OF DESORPTION AT DIFFERENT TEMPERATURES
KD/hr
Period Temperature (°C) b' w Observed Predicted
9
First hour 19.5 2.653 x 10 0.331 0.318
9
Second hour 16.8 3.151 x 10 0.278 0 269
9
Third hour 16 3.314 x 10 0.268 0.256
320
-------
co
ro
TABLE C
SUMMARY DATA OF A DESORPTION EXPERIMENT
Period
(hours)
0-1
1-2
2-3
Initial Final
pH, pl-L
i ^
10.6 10.4
10.4 10.2
10.2 10.1
Temperature °
Initial Final
21.5
17.5
16.0
17.5
16.0
16.0
C
Mean
9.5
16.8
16.0
Ammonia
Initial
Cl
752
554
436
(mgN/1 )
Final
C2
554
436
351
L
h L2
24.4719
24.0651
23.6763
24.0473
23.6677
23.4898
-------
A number of similar experiments were conducted at different temperatures,
different viscosities, different air to liquid ratios, and different liquids
to establish the predictive relationships and the necessary constants which
are described in "Significance of the Research".
322
-------
TABLE VI - 1
BATCH STUDY - NITRIFICATION
OPERATIONAL DATA - UNIT "b"
Day
#
0
1
2
5
6
7
8
9
12
13
14
15
16
19
20
21
22
27
29
TKN
mg/1
350
315
350
298
282
210
199
188
214
219
193
186
183
180
153
173
153
184
171
NH4-N
mg/1
49
126
178
165
146
113
95
94
91
90
87
84
76
76
74
77
74
77
78
N02-N
mg/1
<1
4
12
42
66
110
120
101 .
59
59
13
21
21
21
21
21
21
mg/1
117
78
93
82
99
80
101
131
156
174
170
188
179
189
177
169
N02-N +
mg/1
117
82
105
124
165
190
221
232
215
233
183
188
179
189
177
169
TN
mg/1
467
397
455
422
447
400
420
420
429
452
376
374
362
342
350
340
TN Re-
maining
100
85
97
90
96
86
90
90
92
97
81
80
78
73
75
73
TKN Re-
maining
100
90
100
85
81
60
57
54
61
63
55
53
52
51
44
49
44
53
49
Nitriti-
fi cation
1
1
3
10
15
28
29
24
16
13
4
1
-
-
-
-
-
Nitrati-
fi cation
25
20
23
19
22
20
25
31
36
39
45
50
49
-
55
51
50
Nitrifi-
cation
25
21
26
29
37
48
54
55
52
52
49
51
49
55
51
50
-------
CO
TABLE VI - 2
BATCH STUDY - NITRIFICATION
OPERATIONAL DATA - UNIT "c"
Day
#
0
1
2
5
6
7
8
9
12
13
14
15
16
19
20
21
22
27
29
TKN
mg/1
455
488
385
350
274
223
230
206
219
201
192
194
196
172
164
153
144
175
173
NH4-N
mg/1
43
125
173
188
178
146
109
106
102
101
98
98
92
88
87
87
84
83
83
N02-N
mg/1
<1
3
12
59
72
125
169
163
145
163
155
130
120
27
<1
<1
<1
<1
-------
TABLE VI - 3
BATCH STUDY - NITRIFICATION
OPERATIONAL DATA - UNIT "d"
Day
#
0
1
2
3
6
7
8
9
14
16
20
TKN
mg/1
1103
980
840
516
508
451
403
411
359
298
237
NH4-N
mg/1
205
612
510
394
249
219
186
172
90
56
1
N02-N
mg/1
1
8
15
13
9
6
5
8
19
59
N03-N
mg/1
157
50
67
76
65
44
41
24
25
N02-N +
mg/1
158
58
82
89
74
50
45
32
44
TN
mg/1
1261
1038
922
605
582
501
448
391
342
TN Re-
maining
100
82
73
48
35
40
36
31
27
TNK Re-
maining
100
89
76
47
46
41
39
37
33
27
22
Nitriti-
fi cation
1
1
2
2
2
1
1
2
6
25
Nitrati-
fi cation
14
5
8
15
13
10
10
7
9
-
Nitrifi-
cation
15
6
10
17
15
11
11
9
15
25
-------
CO
ro
TABLE VI - 4
BATCH STUDY - NITRIFICATION
OPERATIONAL DATA - UNIT "e"
Day
#
0
1
2
3
6
7
8
9
14
16
20
TKN
mg/1
1425
1301
T050
718
656
604
569
508
337
411
333
NH4-N
mg/1
269
762
602
500
335
276
202
157
69
50
3
N02-N
mg/1
3
67
54
50
42
29
18
3
8
28
N03-N
mg/1
163
60
99
87
75
57
48
26
32
N02-N +
N03-N
mg/1
166
127
153
137
117
86
66
29
60
TN
mg/1
1591
1428
1203
855
773
690
635
366
393
%
TN Re-
maining
100
90
75
54
49
43
40
23
25
%
TNK Re-
maining
100
91
74
50
46
42
40
36
29
23
%
Nitriti-
f i cati on
2
5
5
6
5
4
3
1
7
%
Nitrati-
f i cati on
10
4
8
10
10
8
6
7
8
%
Nitrifi-
cation
12
9
13
16
15
12
9
8
15
-------
to
ro
TABLE VI - 5
BATCH STUDY - NITRIFICATION
OPERATIONAL DATA - UNIT #40
Day
#
0
1
2
3
4
7
8
9
11
14
16
18
21
23
TKN
mg/1
495
427
384
382
336
256
203
142
158
119
140
133
91
140
NH4-N
mg/1
182
247
227
216
205
142
115
87
62
55
56
52
49
46
N02-N
mg/1
4
9
15
21
44
66
92
131
132
108
114
44
58
N03-N
mg/1
121
113
57
25
32
24
37
46
29
23
50
57
48
N02-N +
N03-N
mg/1
121
117
66
40
53
68
103
138
160
155
158
171
106
TN
mg/1
616
544
450
422
389
324
306
280
318
274
298
304
246
TN Re-
maining
100
89
73
69
63
53
50
46
52
45
48
49
40
TKN Re-
maining
100
86
78
77
68
52
41
29
32
24
28
27
28
Nitriti-
fi cation
1
2
4
5
13
22
33
41
48
36
38
24
Nitrati-
fi cation
20
21
13
6
8
7
12
16
9
8
17
19
20
Nitrifi-
cation
20
22
15
20
13
20
34
49
50
56
53
57
44
continued. . .
-------
TABLE VI-5 -UNIT #40, Continued
Day
#
24
29
31
32
35
36
CO
po 07
oo <3/
39
42
43
45
46
TKN
mg/1
119
91
95
74
70
77
70
49
71
63
63
60
NH4-N
mg/1
45
34
28
27
25
25
22
22
21
21
21
27
N02-N
mg/1
18
1
1
1
1
1
1
1
1
1
1
1
NOs-N
mg/1
61
111
109
101
133
102
93
86
56
85
79
73
N02-N +
N03-N
mg/1
79
111
109
101
133
102
93
86
56
85
79
73
TN
mg/1
198
202
194
175
203
179
163
135
127
148
142
133
%
TN Re-
maining
32
33
32
28
33
29
27
22
21
24
23
22
%
TKN Re-
maining
24
18
19
15
14
16
14
10
14
13
13
12
%
Nitriti-
fi cation
9
1
1
1
-
—
-
-
-
-
-
%
Nitrati-
fication
31
55
53
58
65
57
57
64
44
57
56
55
%
Nitrifi-
cation
40
55
53
58
65
57
57
64
44
57
56
55
-------
TABLE VI - 6
BATCH STUDY - NITRIFICATION
OPERATIONAL DATA - UNIT #50
Day
#
0
1
2
3
4
7
8
9
11
14
16
18
21
23
TKN
mg/1
616
518
441
420
377
273
235
168
154
107
140
140
130
133
NH4-N
mg/1
210
305
257
235
219
135
113
83
52
48
48
42
42
41
N02-N
mg/1
4
8
9
14
38
59
89
151
134
104
107
104
52
N03-N
mg/1
73
73
53
33
42
39
21
37
92
61
73
N02-N +
N03-N
mg/1
81
62
71
101
128
172
171
196
168
125
TN
mg/1
689
521
482
344
336
296
326
278
336
308
258
TN Re-
maining
100
76
70
50
49
43
47
40
49
45
37
TKN Re-
maining
100
72
68
61
44
38
27
25
17
23
23
21
22
Nitriti-
fi cation
1
2
11
18
30
46
48
31
35
20
Nitrati-
fi cat Ion
9
14
11
10
12
13
6
13
27
20
28
Nitrifi-
cation
9
15
13
21
30
43
52
61
58
55
48
continued. . .
-------
TABLE VI-6- UNIT #50, Continued
oo
co
o
Day
#
24
29
31
32
35
36
37
39
42
43
45
46
TKN
mg/1
137
84
81
88
81
-
84
70
71
71
71
67
NH4-N N02-N
mg/1 mg/1
41 44
27 <1
20 <1
18 <1
15 <1
15 <1
14 <1
14 <1
12 <1
13 <1
13 <1
13 <1
N03-N
mg/1
77
131
129
125
135
107
101
89
64
69
77
N02-N +
N03-N
mg/1
121
131
129
125
135
107
101
89
64
69
77
TN
mg/1
258
215
210
213
216
185
169
135
140
144
TN Re-
mai ni ng
37
31
31
31
31
27
25
20
20
21
TKN Re- Nitriti-
maining fi cation
22 17
14 <1
13 <1
14 <1
13
-
14
11
12
12
12
11
Nitrati-
fi cation
30
61
61
59
63
55
56
48
50
54
Nitrifi-
cation
47
61
61
59
63
55
56
48
50
54
-------
CO
CO
TABLE VI - 7
BATCH STUDY - NITRIFICATION
OPERATIONAL DATA - UNIT #60
Day
0
1
2
3
4
7
8
9
11
14
16
18
21
23
TKN
mg/1
785
651
567
532
468
385
333
308
217
168
175
168
168
175
NH4-N
mg/1
305
374
325
308
260
182
164
144
71
41
39
29
32
32
N02-N
mg/1
-
5
9
7
9
27
36
55
134
159
117
150
142
146
N03-N
mg/1
65
97
83
55
40
40
50
19
40
92
49
36
N02-N +
N03-N
mg/1
65
102
92
62
67
76
105
153
199
209
199
182
TN
mg/1
850
748
650
587
425
373
358
236
208
267
217
211
TN Re-
maining
88
77
69
50
44
42
28
25
31
26
25
TKN Re-
maining
83
72
68
60
49
42
28
21
22
21
21
22
Nitriti-
fi cation
-
1
1
1
6
9
13
36
43
31
41
41
Nitrati-
fi cation
8
13
13
9
9
10
12
5
11
24
13
10
Nitrifi-
cation
8
13
14
10
15
19
25
41
54
55
54
51
continued.
-------
TABLE VI-7- UNIT #60, Continued
CO
CO
Day
#
24
29
31
32
35
36
37
39
42
43
45
46
TKN
mg/1
175
140
133
133
105
109
112
91
105
119
109
116
NH4-N N02-N
mg/1 mg/1
31 134
22 52
17 <1
14 <1
6 <1
6 <1
1 <1
1 <1
1 <1
1 <1
1 <1
<1 <1
mg/1
50
115
143
133
151
129
105
93
89
109
88
N02-N +
N03-N
mg/1
184
167
143
133
151
129
105
93
89
109
88
TN
mg/1
225
255
276
266
256
238
217
184
194
228
204
TN Re-
maining
27
30
33
31
30
28
26
22
23
27
24
TKN Re- Nitriti-
maining fi cation
22 37
18 17
17 <1
17 <1
13 <1
14
14
12
13
15
14
15
Nitrati-
fi cation
14
38
52
50
59
54
48
51
46
48
43
Nitrifi-
cation
51
55
52
50
59
54
48
51
46
48
43
-------
CO
GO
GO
TABLE VI - 8
BATCH STUDY - NITRIFICATION
OPERATIONAL DATA - UNIT #70
Day
#
0
1
2
3
4
7
8
9
11
14
16
18
21
23
TKN
mg/1
890
736
622
622
553
406
378
361
291
175
154
147
116
112
NH4-N
mg/1
324
410
359
349
309
221
207
181
139
59
17
4
3
3
N02-N
mg/1
-
7
10
9
6
9
5
11
22
62
101
136
117
125
N03-N
mg/1
65
77
36
51
55
54
17
40
132
97
57
N02-N +
N03-N
mg/1
65
87
45
60
60
65
39
102
233
233
182
TN
mg/1
985
699
658
457
433
415
308
215
286
244
169
TN Re-
maining
100
73
69
48
45
44
32
23
30
26
18
TKN Re-
maining
100
83
70
70
62
46
43
41
33
20
17
17
13
13
Nitriti-
fi cation
0
1
1
1
2
1
3
7
22
26
36
43
V
h
Nitrati-
fi cation
7
11
5
11
13
13
5
14
34
26
19
Nitrifi-
cation
7
12
6
13
14
16
12
36
60
62
62
continued. . .
-------
TABLE VI-8-UNIT #70, Continued
co
co
Day
#
24
29
31
32
35
36
37
39
42
43
45
46
TKN
mg/1
98
98
95
135
105
121
119
119
119
147
133
136
NH4-N
mg/1
1
<1
0
0
0
0
0
0
0
0
0
0
N02-N
mg/1
125
114
90
80
80
17
2
<1
<1
<1
<1
<]
N03-N
mg/1
44
55
55
58
103
89
73
65
73
77
85
N02-N +
N03-N
mg/1
169
169
145
138
183
106
75
65
73
77
85
TN
mg/1
142
153
150
193
208
210
192
184
192
224
221
TN Re-
maining
15
16
16
20
22
22
20
19
20
24
23
TKN Re-
maining
11
11
11
15
12
14
13
13
12
17
15
Nitriti-
fi cat ion
47
43
38
29
28
8
1
<1
<1
-
-
Nitrati-
fication
17
21
23
21
36
39
38
35
38
34
39
Nitrifi-
cation
64
64
61
50
64
47
39
35
38
34
39
-------
CO
en
TABLE VII-1
OBSERVED AND THEORETICAL VALUES OF COD FOR THE OBSERVED N02~N AND NO
DENITRIFICATION RUN V
3-
N DECREASE AT 20°C AND 35°C
OBSERVED
At
(days)
0-.25
.25-1
1-2
2-3
3-6
AN02-N
(mg/1)
20°C 35°C
28 40
42 110
50 60
80 110
120
AN03-N
(mg/1)
20°C
15
5
1
1
1
35°C
15
5
34
7
0
ACOD
(mg/1)
20°C 35°C
_ _
200 550
250 250
-100 -100
-100 -100
ZACOD
(mg/1)
20°C 35°C
_
200 550
450 800
550 900
650 1000
THEORETICAL
At
(days)
0-.25
.25-1
1-2
2-3
3-6
Atheoretical
20°C 35°C
64 92
97 253
115 1 38
184 253
276
COD decrease due to
decrease
20°C
67
19
4
4
4
AN
35°C
35
19
126
26
0
total COD
decrease in the
time period due
to N02-N and
N03-N
20°C 35°C
131 148
116 272
119 264
188 279
280
Atheoretical
COD
20°C 35°C
131 148
247 420
366 684
554 963
834 963
-------
TABLE VII-2
THEORETICAL COD DECREASE DUE TO OBSERVED
N02-N DECREASE - DENITRIFICATION RUN V
At COD decrease (theoretical)
days 20°C 35°C
0-.25 64 92
.25-1 163 345
1-2 278 483
2-3 462 736
3-6 738 736
336
-------
TABLE VIII-1
COMPUTATIONS OF THE AMOUNT OF GLUCOSE AND
CHICKEN MANURE BASED ON THEIR THEORETICAL
COD DENITRIFICATION RUN VII
NO.,-N concentration of Unit F' mixed liquor = 322 mg/1
Glucose Computations
270 mg/1 N03~N needs 1000 mg/1 of COD, therefore
322 mg/1 = - x 322 = 1185 mg/1 of COD
since 192 mg of COD = 180 mg of glucose
1185 mg = ||| x 1185 = 1.111 gm of glucose per liter
actual amount added to Reactor "B" 1-1726 g/1 (Ix)
actual amount added to Reactor "C" 2-3452 g/1 (2x)
Chicken Manure Computations
Amount of COD needed from above =1185 mg/1
assume 1 mg of COD = 1.3 mg of dry total solids of chicken manure
1185 mg/1 of COD = 1.3 x 1185 = 1540.5 mg/1 of total solids
assuming that 1/4 of the COD is only available for denitrification,
we need = 1540.5 x 4 = 6162 mg/1 of total solids
since ^75% of the chicken manure is water (25% solids),
we need 6162 x 4 = 24.648 gm of raw manure (-25 g/1)
actual amount added to Reactor "D" = 25.7 g/1 (CM Ix)
actual amount added to Reactor "E" = 51.4 g (CM 2x)
337
-------
TABLE VII1-2
COMPUTATIONS OF THE AMOUNT OF GLUCOSE AND
CHICKEN MANURE BASED ON THEIR THEORETICAL
COD - DENITRIFICATION RUN VIII
Average NO,,-N concentration of Unit G1 mixed liquor = 885 mg/1
Glucose Computations
theoretical: 435 mg/1 NOp-N needs 1000 mg/1 COD, therefore
885 mg/1 = - x 885 = 2034 m9 of COD
since 192 mg/1 COD = 180 mg/1 of glucose
1 80
2034 mg/1 = 4- x 2034 = 1907 mg of glucose per liter
actual amount of glucose added = 2.0 g (B1 Reactor)
actual amount of glucose added = 4.0 g (C1 Reactor)
Chicken Manure Computations
Amount of COD needed 2034 mg
assume 1 mg of COD = 1.3 mg of dry total solids
2034 mg/1 COD = 1.3 x 2034 = 2644 mg/1 of dry solids
assuming that 1/4 of the COD is only available for biological denitrifi cation,
we need 2644 x 4 = 10.576 g/1 of dry solids
since ~75% of chicken manure is water (~25% dry solids), we need
10.576 x 4 = 42.304 g/1 of wet manure
actual amount added to Reactor "D" = 32.86 g/1
actual amount added to Reactor "E" = 65.7 g/1
338
-------
EVALUATION OF ANALYTICAL METHODS TO DETERMINE
NH4-N, N02-N, AND NOg-N IN POULTRY MANURE
A study was undertaken to examine the efficacy of some of the available
methods for the determination of NH4-N, N02~N, and N03~N contained in the
poultry manure wastewaters.
Methods Studied
NH4-N: a) Method described in Standard Methods (12) - distillation
of the buffered sample (pH 7.4) followed by titration with potassium biiodate,
b) Method described in Methods of Soil Analysis (193) - distillation of the
sample treated with MgO, followed by titration with potassium biiodate.
N02-N: Diazotization method - sulfanilic acid treatment followed by
the addition of N-l, napthyl ethylene diamine dihydrochloride (192).
N03-N: a) method using a specific NOo-ion electrode fitted on an
Orion meter, b) modified PDSA method as described in Materials and Methods
section, c) Brucine method described in Standard Methods (12), d) slightly
modified procedure of the method described in Methods in Soil Analysis (193)
using Devarda's alloy. In this method, distillation of the Devarda's alloy
treated sample gives the results for the total of nitrite and nitrate nitro-
gen. The value of NO?-N obtained by the diazotization method was deducted
to obtain the value for N03-N.
The modified procedure is as follows. A known volume (5 ml) of the
poultry manure sample was taken directly into the microKjeldahl distillation
assembly. To this, 0.5 g of MgO was added. The steam distillation of the
sample was carried out and 30 ml of the distillate was collected into boric
acid and titrated with standard potassium biiodate (1 ml = 0.1401 mg of N)
to obtain the NH4-N concentration in the sample.
339
-------
To the mixture contained in the distillation assembly, 1.5 g of De-
varda's alloy was added and distillation was resumed. The distillate
(30 ml) was collected into boric acid solution and titrated against stan-
dard potassium biiodate to obtain the N02- + N03-N. In a separate exper-
iment, N02-N was determined by the diazotization method. This value was
deducted from the value obtained for N02-N + N03~N in the above determina-
tion to obtain the NCL-N concentration.
General observations
The results obtained for NH4~N according to the distillation and
titration method described in Standard Methods were in agreement with the
results obtained by the distillation method using MgO (range ±4%).
The determination of NCL-N with the Orion meter fitted with an electrode
specific for the nitrate ion was found to be unsatisfactory.
The Brucine method for the determination of NO^-N gave variable results.
However, in several instances, the results agreed within ±10% of the results
obtained by the PDSA method.
The results obtained for N02- + N03-N by the Devarda alloy distilla-
tion method were within ±3-8% of the total N03-N and N02-N obtained by
adding the N03-N and N02-N determined separately by the PDSA and diazotiza-
tion methods respectively.
340
-------
1
Accession Number
w
5
Organization
*-j Subject f* ic Id & Group
05D
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
Agricultural Waste Management Program, College of Agriculture and
Life Sciences, Cornell University, Ithaca, New York 14850
Title
Development and Demonstration of Nutrient Removal from Animal Wastes
i Q Authors)
R.C. Loehr
T.B.S. Prakasam
E.G. Srinath
Y.D. Joo
1A Project Designation
10 EPA/ORM Projects No. 13040 DPA and 13040 DDG
2] Note
04/72
on Citation
number, EPA-R2-73-095, January 1973.
23
Descriptors (Starred First)
*Nitrogen Control, *Phosphorus Control, Nitrification, Denitrification,
Ammonia Stripping, Chemical Precipitation, Predictive Relationships,
Animal Wastes
25
Identifiers (Starred First)
*Nutrient Control, *Animal Waste Treatment Processes
27
Abstract
Laboratory and pilot plant studies evaluated the feasibility of a) chemical pre-
cipitation, b) ammonia removal by aeration, and c) nitrification and denitrifi-
cation as methods to remove nitrogen, phosphorus, and color from animal waste-
waters. Poultry and dairy manure solutions were used over a broad concentration
range to illustrate the fundamentals of the processes as applied to these wastes
and to demonstrate the applicability of the processes.
Alum, lime, and ferric chloride can be used for phosphorus control in animal waste-
water although the chemical costs are from 2-10 times those quoted for municipal
wastewater. Two predictive relationships were determined that appear useful for
design and operation of phosphate removal systems.
Ammonia removal was found to be feasible and detailed equations were developed and
verified to determine the ammonia loss under specific environmental conditions.
Nitrification followed by denitrification was found to be technically feasible.
Parameters affecting the design and performance of these processes with animal
wastewaters were identified.
Abstn
R.C. Loehr
Institution
Cornell University
WR:102 (REV JULY 1969
SEND WITH COPY OF DOCUMENT, TO: WATER RESOURCES SCIENTIFIC INFOF
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON. D. C. 20240
U.S GOVERNMENT PRINTING OFFIC£:1973 514-152/163 1-3
------- |