USF Libraries
USF Digital Collections

Stimulation of nitrification by carbon dioxide in lab-scale activated sludge reactors

MISSING IMAGE

Material Information

Title:
Stimulation of nitrification by carbon dioxide in lab-scale activated sludge reactors
Physical Description:
Book
Language:
English
Creator:
Posso-Blandon, Lina
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
Ammonium
Nitrate
Nitrifying bacteria
Nitrifiers
Wastewater
Dissertations, Academic -- Environmental Engineering -- Masters -- USF   ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: Wastewater treatment plants (WWTPs) are required to remove ammonium (NH4+) from wastewater due to its oxygen demand and toxicity to the aquatic organisms. Ammonium is removed in the activated sludge treatment system by nitrification and denitrification processes. Nitrification is the oxidation of NH4+ to nitrate (NO3-) by autotrophic nitrifying bacteria which use carbon dioxide (CO2) as a carbon source for growth. These bacteria grow slowly with low nitrification rates limiting WWTPs capacity. In this research it was hypothesized that supplying higher concentrations of CO2 during aeration increases nitrification rates, resulting in a reduction of the solids retention time (SRT).
Thesis:
Thesis (M.S.E.V.)--University of South Florida, 2005.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Lina Posso-Blandon.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 91 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001670409
oclc - 62367646
usfldc doi - E14-SFE0001285
usfldc handle - e14.1285
System ID:
SFS0025606:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001670409
003 fts
005 20051216093401.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 051129s2005 flu sbm s000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001285
035
(OCoLC)62367646
SFE0001285
040
FHM
c FHM
049
FHMM
090
TA170 (Online)
1 100
Posso-Blandon, Lina.
0 245
Stimulation of nitrification by carbon dioxide in lab-scale activated sludge reactors
h [electronic resource] /
by Lina Posso-Blandon.
260
[Tampa, Fla.] :
b University of South Florida,
2005.
502
Thesis (M.S.E.V.)--University of South Florida, 2005.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 91 pages.
3 520
ABSTRACT: Wastewater treatment plants (WWTPs) are required to remove ammonium (NH4+) from wastewater due to its oxygen demand and toxicity to the aquatic organisms. Ammonium is removed in the activated sludge treatment system by nitrification and denitrification processes. Nitrification is the oxidation of NH4+ to nitrate (NO3-) by autotrophic nitrifying bacteria which use carbon dioxide (CO2) as a carbon source for growth. These bacteria grow slowly with low nitrification rates limiting WWTPs capacity. In this research it was hypothesized that supplying higher concentrations of CO2 during aeration increases nitrification rates, resulting in a reduction of the solids retention time (SRT).
590
Adviser: Peter G. Stroot.
653
Ammonium.
Nitrate.
Nitrifying bacteria.
Nitrifiers.
Wastewater.
690
Dissertations, Academic
z USF
x Environmental Engineering
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.1285



PAGE 1

Stimulation of Nitrification by Carbon Dioxide in Lab-Scale Activated Sludge Reactors by Lina Posso-Blandon A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Environmental Engineering Department of Civil and Environmental Engineering College of Engineering University of South Florida Major Professor: Peter G. Stroot, Ph.D Daniel Yeh, Ph.D Jeffrey Cunningham, Ph.D Date of Approval: July 20, 2005 Keywords: ammonium, nitrate, nitrifying bacteria, nitrifiers, wastewater Copyright 2005 Lina Posso-Blandon

PAGE 2

Acknowledgements I would like thank my major professor, Dr. Peter G. Stroot for all his guidance and support during the course of my masters. His interest in environmental problems and his dedication to research is inspiring. I would also like to thank him for providing the resources to conduct this research. I appreciate the help Dr. Jeffrey Cunni ngham and Dr. Daniel Yeh provided, as member of my thesis committee. Their comments and suggestions helped me to finish this document. Thanks to Ivan Zapata from Dr. Stroots Lab for helping me with the system set up, data collection, and daily supply of water. Thanks to Matt Cutter from Dr. Stroots Lab for helping me with data collection and for collaborating with and reviewing my written document. Thanks to the Department of Civil and Environmental Engineering for providing me with the resources to support myself through my masters. Thanks to my friends Danielly Orozco, Ashutosh Vakharkar, and Carolina Marcos for their help during tough moments. Finally, thanks to my parents and brothers for their love and support.

PAGE 3

i Table of Contents List of Tables................................................................................................................. ....iv List of Figures................................................................................................................ .....v Abstract....................................................................................................................... ......vii 1. Introduction................................................................................................................ ...1 2. Literature Review.........................................................................................................3 2.1 The Activated Sludge System.................................................................................3 2.1.1 Sequencing Batch Reactors.......................................................................4 2.2 Biological Nutrient Removal..................................................................................5 2.3 Nitrification 6 2.3.1 Stoichiometry............................................................................................6 2.3.2 Microbiology.............................................................................................7 2.3.3 Factors Affecting Nitrification Performance.............................................9 2.3.4 Effect of CO2 on Nitrifying Bacteria.......................................................10 3. Hypotheses.................................................................................................................. 12 3.1 Central Hypothesis................................................................................................12 3.1.1 Hypothesis 1............................................................................................12 3.1.2 Hypothesis 2............................................................................................12 4. Objectives.................................................................................................................. .13 5. Materials and Methods................................................................................................14 5.1 System Configuration...........................................................................................14 5.1.1 Air Supply System...................................................................................15 5.1.2 System Configuration..............................................................................16 5.2 Experimental Design.............................................................................................18 5.2.1 Solids Retention Time.............................................................................19

PAGE 4

ii 5.2.2 Supply of Air and CO2 to the Experimental Reactor..............................20 5.2.3 Water.......................................................................................................20 5.3 Data Collection and Sample Analyses..................................................................21 5.3.1 Analytical Methods.................................................................................23 6. Model of the Impact of pCO2 on pH...........................................................................26 7. Results and Discussion...............................................................................................30 7.1 Preliminary Results...............................................................................................30 7.2 Experiment 8: Supply of CO2 during the Full React Cycle..................................30 7.2.1 Nitrate Formation Rates Exp. 8............................................................30 7.2.2 Settling Performance Exp. 8.................................................................33 7.2.3 COD Removal Exp. 8...........................................................................35 7.2.4 Summary of Results Exp. 8..................................................................36 7.3 Experiment 9: Supply of CO2 during the Last 5 Hours of the React Cycle.........36 7.3.1 Nitrate Formation Rates Exp. 9............................................................36 7.3.2 Settling Performance Exp. 9.................................................................38 7.3.3 COD Removal Exp. 9...........................................................................40 7.3.4 Summary of Results Exp. 9..................................................................41 7.4 Experiment 10: Tripled SRT Confirmatory Experiment...................................41 7.4.1 Nitrate Formation Rates Exp. 10..........................................................42 7.4.2 Total and Volatile Suspended Solids Exp. 10......................................43 7.4.3 Specific Nitrate Formation Rates Exp. 10............................................47 7.4.4 Settling Performance Exp. 10...............................................................50 7.4.5 Ammonium Removal Exp. 10..............................................................53 7.4.6 Nitrate Concentrations in Supernatant Exp. 10....................................55 7.4.7 COD removal..........................................................................................57 7.4.8 Summary of Results................................................................................59 7.5 Future Research....................................................................................................61 8. Conclusions................................................................................................................. 63 9. References.................................................................................................................. .65 10. Bibliography............................................................................................................... 70

PAGE 5

iii Appendices........................................................................................................................71 Appendix A: Calibration of the Ion Selective Electrodes...........................................72 A.1. Calibration of the Ammonium Probe.........................................................73 A1.2. Calibration of the Nitrate Probe.................................................................77

PAGE 6

iv List of Tables Table 1. Bacteria Classification Acco rding to the Carbon and Energy Source .................7 Table 2. Typical Parameters for Heterotrophic and Autotrophic Nitrifying Bacteria. ......8 Table 3. Composition of Synthetic Wastewater ..............................................................16 Table 4. Quality of Synthetic Wastewater .......................................................................17 Table 5. SBR Operational Parameters .............................................................................17 Table 6. Experimental Design .........................................................................................19 Table 7. Data and Sample Collection Strategy ................................................................23 Table 8. Molecular Weight and Concentration of CaCO3 and NaHCO3 in eq/mole and mg/meq .....................................................................................................28 Table 9. Typical Alkalinity Values in Wastewater and their Correspondent Concentrations of Na+ -when Adjusting Alkalinity with NaHCO3 -...............28 Table 10. Average Solids Concentrations and Biomass Content Experiment 10 ..........47 Table 11. Specific Nitrate Formation Rates during Different SRTs Exp.10 .................49 Table A-1. Ions Present in the Synthetic Wastewater .....................................................72 Table A-2. Standard Solutions for the Calibration of the Ammonium Probes ................73 Table A-3. Feed Samples with Variable Nitrate Concentrations for the Calibration of the Ammonium Probes ...................................................................................74 Table A-4. Calibration of Ammonium Probes with Standard Solutions (Day 2) ............76 Table A-5. Ammonium Concentration in Feed Samples during Calibration (Day 2) .....77 Table A-6. Standard Solutions for the Calibration of the Nitrate Probes ........................78 Table A-7. Feed Samples with Variable Nitrate Concentrations for the Calibration of the Nitrate Probes ............................................................................................78 Table A-8. Calibration of Nitrate Probes with Standard Solutions (Day 7) ....................80 Table A-9. Nitrate Concentrations in Feed Samples during Calibration (Day 7) ...........81

PAGE 7

v List of Figures Figure 1. Typical Configuration of an Activated Sludge System......................................3 Figure 2. Activated Sludge System Oper ated as a Sequencing Batch Reactor.................4 Figure 3. Dessicator Cabinet (A) and Chamber after Modifications (B).........................14 Figure 4. Air and CO2 Supply System.............................................................................15 Figure 5. Activated Sludge SBR System for Nitrification with pCO2 Control...............18 Figure 6. Sludge Blanket Volume for Settling Performance Preliminary Test...............22 Figure 7. Filtration Apparatus for SS Analyses...............................................................24 Figure 8. pH as a Function of pCO2 and Alkalinity.........................................................29 Figure 9. Nitrate Formation Day 3, Experiment 8.........................................................31 Figure 10. Nitrate Formation Rates Experiment 8.........................................................32 Figure 11. Sludge Blanket Volume per 100 mL of Sample Experiment 8....................34 Figure 12. COD Removal Efficiencies Experiment 8...................................................35 Figure 13. Nitrate Formation Rates Experiment 9.........................................................37 Figure 14. Sludge Blanket Volume per 100 mL of Sample Experiment 9....................39 Figure 15. COD Removal Efficiencies Experiment 8...................................................40 Figure 16. Nitrate Formation Rates Experiment 10.......................................................42 Figure 17. Total Suspended Solids Experiment 10........................................................45 Figure 18. Volatile Suspended Solids Experiment 10...................................................45 Figure 19. Percentage content of Volatile Suspended Solids Experiment 10...............46 Figure 20. Specific Nitrate Formation Rates Experiment 10.........................................48 Figure 21. Sludge Volume Index in Both Reactors Experiment 10..............................51 Figure 22. Sludge Blanket Test for Both React ors Day 34, Experiment 10. Arrows indicate fragmented sludge blanket................................................................52 Figure 23. Removal Efficiency of Ammonium Experiment 10....................................54

PAGE 8

vi Figure 24. Average Ammonium Concentration in Supernatant for different SRTs during Experiment 10.................................................................................................54 Figure 25. Average Nitrate Concentration in Supernatant Experiment 10...................56 Figure 26. Average Nitrate Concentration in Supernatant per SRT Exp.10.................57 Figure 27. Removal Efficiency of COD Experiment 10...............................................58 Figure 28. COD in the Supernatant Experiment 10......................................................58 Figure A-1. Ammonium Probes Calibration Curves .......................................................76 Figure A-2. Nitrate Probes Calibration Curves ...............................................................80

PAGE 9

vii Stimulation of Nitrification by Carbon Dioxide in Lab-Scale Activated Sludge Reactors Lina Posso-Blandon Abstract Wastewater treatment plants (WWTPs) are required to remove ammonium (NH 4 + ) from wastewater due to its oxygen demand and toxicity to the aquatic organisms. Ammonium is removed in the activated sludge treatment system by nitrification and denitrification processes. Nitrification is the oxidation of NH 4 + to nitrate (NO 3 ) by autotrophic nitrifying bacteria which use carbon dioxide (CO 2 ) as a carbon source for growth. These bacteria grow slowly with low nitrification rates limiting WWTPs capacity. In this research it was hypothesized that supplying higher concentrations of CO 2 during aeration increases nitrification rates, resulting in a reduction of the solids retention time (SRT). This hypothesis was tested with two lab-scale sequencing batch reactors seeded with sludge from a full-scale activated sludge WWTP and fed synthetic wastewater. The control reactor was aerated with regular air (0.03% CO 2 ) and the experimental reactor was aerated with air containing 1% CO 2 Ammonium and NO 3 were measured online to determine the nitrification rates. Samples for solids and chemical oxygen demand (COD) determination were collected to evaluate the system performance. Supplying CO 2 to the experimental reactor throughout the entire react cycle resulted in proliferation of filamentous bacteria, poor settling, and washout of the biomass. However, nitrate formation rates in the experimental reactor were 3 times higher than the control before washout occurred. In a subsequent experiment, CO 2 was

PAGE 10

viii supplied to the experimental reactor only during the last 5 hours of the react cycle, resulting in excellent settling and nitrification rates 6 times higher than in the control. A confirmatory experiment was conducted that lowered the SRT from 8 days to 6, 4, and 2 days. Nitrate formation rates were up to 12 times higher in the experimental reactor compared to the control, with an average of 4 times higher. Additionally, the sludge volume index (SVI) suggested a positive impact of CO 2 on settling performance. No impact of CO 2 on COD removal was observed. The results obtained suggest a positive effect of CO 2 on the nitrate formation rates and settling performance in the activated sludge system, indicating that nitrification can be achieved at low SRTs which might optimize WWTPs capacity.

PAGE 11

1 1. Introduction Currently in the United States, 32 billion gallons of municipal wastewater are treated effectively every day (Metcalf & Eddy 2003). Most of the wastewater across the country is treated by wastewater treatment plants (WWTP) that use the Activated Sludge process. In this process, the organic compounds present in the wastewater, usually measured as biochemical oxygen demand (BOD) and/or chemical oxygen demand (COD), are biologically degraded. The activated sludge is the biomass that biologically degrades the organic compounds into CO 2 and new biomass. The average period of time during which the activated sludge remains in the system is called solids retention time (SRT) and is the most critical parameter for adequate treatment performance. The SRT values for BOD removal commonly range from 3 to 6 days. Longer periods are required for lower wastewater temperatures (Metcalf & Eddy 2003). Although removal of BOD and total suspended solids (TSS) is the main goal of the activated sludge treatment, in recent years the process has evolved by incorporating the biological removal of nitrogen and phosphorus. Removal of nitrogen is accomplished by nitrification and denitrification processes. For instance, in the nitrification process ammonium (NH 4 + ) is oxidized to nitrate (NO 3 ). Nitrate is subsequently reduced to nitrogen gas (N 2 ) and removed from the aqueous phase during denitrification. However, nitrification is a slow process that requires a long SRT, as high as 20 days, which might limit WWTPs capacity. An increasing U.S. population requires a larger capacity from WWTPs, which can be achieved either by expansion or by faster treatment processes. Expansion of existent WWTP raises capital and operational costs. On the other hand, accelerating the treatment process can augment the WWTP capacity cost-effectively.

PAGE 12

2 Therefore, alternate process configurations may need to be incorporated into the current activated sludge WWTP designs to meet future demands and the removal of nitrogen. This research evaluated an alternate treatment in a lab-scale activated sludge process to achieve the removal of NH 4 + operating at a shorter SRT. It was hypothesized that the addition of carbon dioxide (CO 2 ) to the reactor during aeration will result in faster nitrification rates and nitrification c ould be maintained at low SRTs. Even though faster nitrification rates were obtained when CO 2 was supplied could suggest that the growth rate of nitrifying autotrophic bacteria was stimulated by the addition of CO 2 this research did not analyzed the growth rates. However, samples for later biomass analyses were collected and preserved for future research.

PAGE 13

2. Literature Review 2.1 The Activated Sludge System Activated sludge system is a biological process for the secondary treatment of wastewater for the removal of BOD and TSS. A typical configuration of an activated sludge system is depicted in Figure 1. In general, the effluent wastewater from the primary clarifier passes through an aerated, complete-mix tank with mixed-liquor suspended solids (MLSS) that contains a wide variety of microorganisms capable of degrading organic waste. The microorganisms or biomass remain suspended in the aeration tank for an average residence period defined as the hydraulic retention time (HRT) before passing through a quiescent non-aerated basin for removal of suspended solids (SS) by gravity settling. A portion of the settled solids is usually recirculated to the aeration basin to control the solids retention time (SRT). The high concentration of biomass due to recycling of the sludge, allows the liquid detention time or HRT to be small. The wasting of sludge separately from the liquid makes the SRT separate from and much larger than the HRT. Figure 1. Typical Configuration of an Activated Sludge System Return activated slud g e Aeration/reaction basin Secondary clarifier Primary effluen t Secondary effluent Waste activated sludge 3

PAGE 14

The secondary effluent of the activated sludge system is then filtered and disinfected or further treated, depending on its future use. 2.1.1 Sequencing Batch Reactors The two steps of the activated sludge system usually take place in two separate reactors as shown in Figure 1, but they can also occur in sequential cycles within a single batch reactor. This type of system, known as the sequencing batch reactor (SBR), is suitable and convenient for lab-scale experiments. The efficacy of SBRs in providing high le vels of biological nutrient removal in activated sludge systems has been extensively demonstrated worldwide (Peters et al. 2004). Figure 2, below, illustrates the phases of the activated sludge system that take place in a single batch reactor when using th e SBR system. In the SBR configuration, wastewater is added during the filling period to a single reactor where equalization, aeration, and clarification can all be achieved. Once the reactor is full, it behaves like a conventional activated sludge system, but without a continuous influent or effluent flow. The aeration and mixing are discontinued after the react period is complete, the biomass settles, and the treated supernatant is removed. Excess biomass is wasted once per day at the end of the react period of the third cycle, to control the SRT. Sequencing batch reactor systems have been successfully used to treat both municipal and industrial wastewater, and are uniquely suited for wastew ater treatment applications characterized by low or intermittent flow conditions (U.S.EPA 1999). Influen t Ai r 1. Fill 2. React 3. Settle 4. Decan t Figure 2. Activated Sludge System Operated as a Sequencing Batch Reactor 4

PAGE 15

5 2.2 Biological Nutrient Removal The activated sludge system is designed for the removal of the COD and TSS from the wastewater. Removal of these contaminants has evolved since the activated sludge process was first used in the early 20 th century. In response to the need for effluents of higher quality, WWTPs are using di fferent configurations for the activated sludge process that incorporate new technology, better understanding of the microbial processes, and reduction of capital and operational costs. Additionally, to achieve higher removal efficiency of nutrients, several modifications have been introduced, including an anaerobic stage for denitrification, anoxic zones and oxidation ditches, sludge recycling, addition of pure oxygen during aeration, different mixing regimes and tank geometries, the use of membrane bioreactors, and the use of SBRs (Grady et al. 1999). Recent regulations concerning nutrient discharges require nitrogen and phosphate removal (Chapter 62-600 F.A.C. Part III, Domestic Wastewater Facilities: Treatment requirements, Florida Department of Envir onmental Protection, FDEP). The need for nutrient removal arises from water quality concerns over the effects of nutrients on the aquatic environment and water reuse operations. As regulations become more stringent, the incorporation of biological nutrient removal (BNR) has been one of the recent challenges in the activated sludge treatment process. BNR systems are modifications of the activated sludge process that incorporate anoxic and/or anaerobic phases to provide nitrogen and/or phosphorus removal. The aerobic phase is a necessary component of all BNR systems, the anaerobic phase is necessary to accomplish phosphorus removal, and the anoxic phase is necessary for nitrogen removal (Grady et al. 1999). Because of the different characteristics of each phase, SBRs are commonly used for BNR processes. Hence, multiple basins are not required and nutrient removal can be achieved costeffectively. Due to the increasing reuse of treated wastewater in agriculture and industrial applications, special interest has been focused on nitrogen removal through the nitrification/denitrification process in BNR systems.

PAGE 16

2.3 Nitrification The removal of nitrogen from wastewater is desired in order to avoid eutrophication in receiving surface waters, to eliminate ammonium (NH 4 + ) toxicity to water and aquatic organisms, to diminish the large dissolved oxygen (DO) demand exerted by NH 4 + and to provide nitrogen control for water-reuse applications (Rittman and McCarty 2001; Metcalf & Eddy 2003). In the activated sludge process, nitrogen removal is accomplished by two biological processes: nitrification and denitrification. Nitrification is a two-step aerobic process in which bacteria oxidize NH 4 + to nitrite (NO 2 ) and then to nitrate (NO 3 ). This process is followed by denitrification, which is the anaerobic reduction of NO 3 to nitrogen gas (N 2 ) subsequently released to the atmosphere. In the conventional activated sludge system, nitrification can occur along with BOD removal in a single aerated reactor (called single-sludge system) or in separate consecutive basins (two-sludge system) if toxic substances are present in concentrations that inhibit the nitrification process (Rittman and McCarty 2001). In the two-sludge systems, BOD removal is usually accomplished before nitrification takes place. In both cases, denitrification takes place in a different basin or in a different phase if the system is operating as SBRs. 2.3.1 Stoichiometry The nitrification process is achieved by two sequential reactions. The biological oxidation of NH 4 + to NO 2 is shown by Equation 1 below a nd is followed by the oxidation of NO 2 to NO 3 as shown in Equation 2 (Metcalf & Eddy 2003). OH2H4NO2O3NH22 2 24 (1) 3 22NO2O 2NO (2) As shown in Equations 1 and 2, both steps of the nitrification process require oxygen, which suggests DO to be a limiting factor of the reaction. It is also to note that the protons produced in the oxidation reactions cause a reduction in pH. These factors 6

PAGE 17

affecting the nitrification process will be discussed later. In addition to the oxygen required, CO 2 is necessary for biomass synthesis. The overall nitrification reaction also includes the assimilation of the NH 4 + ion into biomass cells. The combination of the reactions for autotrophic cell synthesis, oxidation of NH 4 + to NO 3 and reduction of O 2 to water, is shown in Equation 3 (Metcalf & Eddy, 2003). 1.98HOH09.0NO98.0NOHC02.0CO01.0O86.1NH2 3 275 2 2 4 (3) Therefore, Equation 3 describes the overall nitrification reaction carried out by nitrifying autotrophic bacteria. 2.3.2 Microbiology In addition to the phylogenetic classification of microorganisms, bacteria are also classified according to their source of energy and carbon used for cell synthesis. Table 1 presents the terminology used for some nitrifying bacteria. Table 1. Bacteria Classification According to the Carbon and Energy Source (adapted from Rittman and McCarty 2001) Carbon Source Energy Source Inorganic Carbon Organic Carbon Light Chemical Reactions Chemoorganotrophs 1 Autotrophs Heterotrophs Phototrophs Chemotrophs Chemolithotrophs 2 1 Organic chemicals 2 Inorganic chemicals Two phylogenetically distinct groups of chemolithoautotrophic bacteria are responsible for the two-step oxidation of NH 4 + to NO 3 The nitrosifiers oxidize NH 4 + to NO 2 and the nitrifiers oxidize NO 2 to NO 3 These nitrifying bacteria are autotrophic, which means they use inorganic carbon (e.g. CO 2 ) as their source of carbon for growth, and are also termed chemolithotrophs for their source of energy is the chemical reaction of inorganic compounds. Oxidation of NH 4 + to NO 2 can also be carried out by 7

PAGE 18

8 heterotrophic bacteria. However, these bacteria require high energy to oxidize NH 4 + which is a disadvantage when compared to ammonium-oxidizing autotrophic bacteria. Autotrophic nitrifying bacteria are thought to have slow growth rates and are sensitive to pH and temperature swings, making nitrification difficult to maintain in activated sludge systems (Mobarry et al. 1996; Wagner et al. 1996). Table 2 compares typical growth yield values (Y) and maximum specific growth rates () of aerobic heterotrophic bacteria and NH 4 + and NO 2 oxidizing bacteria at 20C. As shown in Table 2, the maximum specific growth rate, of both nitrosifiers and nitrifiers is low compared to that of hete rotrophic bacteria, with both rates less than 1 d -1 at 20C (Rittman and McCarty 2001). Additionally, compared to aerobic heterotrophic bacteria, nitrifying bacteria synthesize very few electrons of the substrate into biomass (low f s o values shown in Table 2), which results in low Y values. This explains the slow nature of the nitrification process. Theref ore, their slow growth rate slows down the nitrification process requiring a lengthy SRT -a s high as 20 daysto prevent washout of the biomass. The high SRT may limit the capacity of existing WWTPs (Metcalf & Eddy 2003). Table 2. Typical Parameters for Heterotrophic and Autotrophic Nitrifying Bacteria (Rittman and McCarty 2001). Parameter Heterotrophic bacteria Ammonium oxidizers Nitrite oxidizers Growth yield 1 Y 0.49 0.33 0.083 Maximum specific growth rate, (d -1 ) 13.2 0.76 0.81 Portion of electron-donor synthesized into cells, f s o (g cells/mol cells) 0.70 0.14 0.10 1 Units of Y: for heterotrophic bacteria: mg VSS/mg BOD L ; ammonium oxidizers: mg VSS/mg NH 4 + -N; and nitrite oxidizers: mg VSS/mg NO 2 -N. In the nitrification process there are dominant species in the activated sludge systems that perform the oxidation of NH 4 + and NO 2 The oxidation of NH 4 + in the activated sludge system was generally attributed to Nitrosomonas europea (Schramm et al. 1998). However, more than 16 speci es of lithoautotrophic ammonia-oxidizing bacteria have been isolated and describe d (Juretschko et al. 1998; Van Loosdrecht 1998).

PAGE 19

9 Dominance of Nitrosococcus mobilis-like bacteria and Nitrosospira spp. among the ammonium-oxidizing bacteria in activated sludge has been revealed by using the molecular biology identification method, fluorescence in situ hybridization (FISH), which uses rRNA-targeted nucleic acid probes for direct quantitative identification of microbial populations (Head et al. 1993; Wagner et al. 1996; Juretschko et al. 1998; Schramm et al. 1998; Hall et al. 2003) Similarly, the oxidation of NO 2 has been usually attributed to Nitrobacter agilis (Van Loosdrecht 1998). However, several studies using FISH have found that Nitrospira-like bacteria are present in significant numbers in typical activated sludge systems (Juretschko et al. 1998; Schramm et al. 1998; Hall et al. 2003; Kim et al. 2004). Although significan t contributions to the study of the microbiology of nitrification have been made our understanding of nitrifying bacteria is still limited and further research is needed to optimize the nitrification process. 2.3.3 Factors Affecting Nitrification Performance Because of the biological nature of the n itrification process, the performance of the treatment is affected by several environm ental factors including DO, pH, temperature, toxic substances, and metals. One critical factor affecting nitrification is the pH as shown in Equations 1 and 2. Optimal nitrification rates occur at pH values ranging from 7.5 to 8.5, with rates declining significantly at pH below 6.8 (U.S.EPA 1993). Additionally, the growth rate of nitrifying bacteria is sensitive to temperature, making nitrification difficult to maintain at low SRTs when temperatures are low. Another critical factor is the availability of dissolved oxygen, as shown in Equations 1 and 2. A DO concentration greater than 1.1 mg O 2 /L is usually required for adequate performance of the nitrification process (Rittman and McCarty 2001; Martins et al. 2003). Noda et al. (2003) reported nitrification efficiencies as low as 36% caused by a low DO concentration (0.3 mg O 2 /L), compared to 95% efficiencies when the DO concentration was greater than 2 mg O 2 /L. These results suggested a reduction in the activity and/or quantity of nitrifying bacteria caused by the insufficient oxygen supply. A s econdary effect of low DO concentrations is the potential growth of filamentous bacteria, which usually affects settling performance, and, consequently, the effluent quality. Growth of filamentous bacteria is

PAGE 20

10 thought to be favored by BNR processes although most of these bacteria are still unidentified and the factors favoring its proliferation are not well understood (Liu et al. 2001). Current research has reported proliferation of filamentous bacteria in systems with low DO concentrations and positive effects of inorganic carbon (i.e. HCO 3 ) to remedy poor settling due these bacteria (Wett et al. 2003). However, further research is still needed to determine the relation between growth of filamentous bacteria and pCO 2 (Gaval et al. 2002). In general, the effect of inorganic carbon on the nitrification process is yet to be determined. 2.3.4 Effect of CO 2 on Nitrifying Bacteria Previous work has demonstrated that the growth of some autotrophic bacteria is carbon limited (Dagley and Hinshelwood 1938; Green et al. 2002; Denecke and Liebig 2003). For instance, inorganic carbon was found to be a limiting factor in BNR aerated systems due to the low partial pressure of CO 2 of the atmospheric air being introduced, and to the CO 2 stripped to the atmosphere caused by bubbling (Wett and Rauch 2003). These particular factors were reported to limit the concentration of CO 2 in wastewater and consequently to affect nitrification. Moreover, Wett and Rauch (2003) suggest that pH is not a limiting factor itself, but instead the limiting factor seems to be the bicarbonate limitation resulting from a low pH. Additional influence of CO 2 in the growth rate of nitrifying bacteria has been demonstrated in a chemostat with CO 2 concentrations of up to 17% (Denecke and Liebig 2003). These preliminary results suggested a strong influence of CO 2 on nitrification rates. However, further research is needed in order to determine the influence of SRT on the nitrification process, to determine the impact of higher concentrations of CO 2 on the overall activated sludge system performance, and to identify the CO 2 -sensitive bacteria. Green et al. (2002) found a correlation between the concentration of CO 2 and the ammonium oxidation rate on a nitrifying chalk reactor. In this experiment, the oxidation rate of ammonium increased as the CO 2 concentration was raised. Hence, it was found that the CO 2 concentration limited the nitrification rates up to 0.3 mmol/L (1% pCO 2 ). However, CO 2 concentrations higher than 1% did not affect nitrification.

PAGE 21

11 Other researchers have found that CO 2 stimulates nitrification in the soil. Carbon dioxide is usually found in the soil at a pCO 2 of 10 -2 (1% CO 2 ). Kinsbursky and Saltzman (1990) reported CO 2 as a possible limiting substrate for nitrifying bacteria in the soil.

PAGE 22

12 3. Hypotheses 3.1 Central Hypothesis The growth of autotrophic nitrifying bacter ia in activated sludge systems is limited by the availability of inorganic carbon. 3.1.1 Hypothesis 1 Nitrification is possible at a low SRT in ac tivated sludge systems when air is supplied with high concentrations of CO 2 3.1.2 Hypothesis 2 Nitrifying bacteria have higher nitrate formation rates when provided with air containing higher concentrations of CO 2

PAGE 23

13 4. Objectives The purpose of this research is to evaluate the effect of CO 2 on nitrification rates in the activated sludge process. Five particular objectives are addressed in this project as follows: 1. Design and set up a lab-scale reactor system that features pCO 2 control. 2. Develop a model to predict the impact of pCO 2 on the wastewater pH. 3. Evaluate the performance (i.e. COD removal and solids settling) of a lab-scale activated sludge system when high concentrations of CO 2 are supplied during aeration. 4. Evaluate the effect of CO 2 addition on nitrification rates in the lab-scale reactor system. 5. Compare the nitrification rates at differ ent SRTs for two systems aerated with atmospheric and elevated CO 2 concentrations.

PAGE 24

5. Materials and Methods 5.1 System Configuration A lab-scale reactor system with two SBRs for the treatment of activated sludge, featuring control of pCO 2 was designed and fabricated. The control reactor was aerated with regular air (0.03% CO 2 ) and the experimental reactor was supplied with air with a CO 2 concentration of 1%. Air to the control react or was introduced from the room atmosphere while the experimental reactor was supplied with air from a chamber with a controlled concentration of 1% CO 2 An acrylic dessicator cabinet (Fisherbrand* Acrylic Dessicator, Fisher Scientific, Pittsburg, PA), shown in Fi gure 3(A), was used as the chamber to confine the experimental reactor in the CO 2 controlled atmosphere. The dessicator was modified as shown in Figure 3(B) to set in the equipment for air and CO 2 supply, the apparatus to control the CO 2 concentration within the chamber, and the tubing for feeding and wasting of the reactor. A B CO2sensor Air and CO2 supply Feed tubing Waste tubing Power cords Perforations for probes Figure 3. Dessicator Cabinet (A) and Chamber after Modifications (B) 14

PAGE 25

5.1.1 Air Supply System The air supply system included the addition of CO 2 to the experimental reactor. This approach was designed for this research and has not been used before in lab-scale SBR reactors for activated sludge treatment. The CO 2 concentration inside the chamber was maintained at 1.0% 0.1% using a CO 2 sensor and controller (CARBOCAP AC100 Carbon Dioxide Sensor, Coy Laboratory Products Inc, Grass Lake, MI). The CO 2 was injected from a gas cylinder (Carbon Dioxide Airgas Inc., Randor, PA) to the chamber and regulated by the CO 2 controller which used a solenoid valve to keep the set CO 2 concentration within the chamber. Regular air was also supplied to the chamber and mixed with the CO 2 gas using a computer fan. The mixed air with a concentration of 1% CO 2 was pumped into the experimental reactor. The control reactor was aerated directly with room air using an air pump (OPTIMA Air Pump, Rolf C. Hagen U.S.A. Corp, Mansfield, MA). Two more identical OPTIMA air pumps were used to introduce regular air into the chamber and to introduce the mixed air from the chamber into the experimental reactor. The air injected into the reactors was dispersed with aquarium air stones. Figure 4 illustrates the air supply system for both reactors. Figure 4. Air and CO 2 Supply System Control Reactor Experimental Reactor 1.0%CO2 Air stones CO2 Sensor Chambe r Air Pump CO2 CO2 from Gas c y linde r CO2Controlle r Air stones Air Pump 15

PAGE 26

5.1.2 System Configuration The reactors had a total volume of 3 L and operated as SBRs. Both reactors were seeded with 1 L of sludge from the nitrification basin of a full-scale activated sludge WWTP operated at an SRT of 22 days. Both reactors were fed 2 L of synthetic wastewater every cycle. Table 3 presents the composition of the synthetic wastewater produced daily in the laboratory to feed the reactors with the organic carbon, nutrients, traces metals, and alkalinity typical of municipal wastewater. Table 3. Composition of Synthetic Wastewater 1 Added when dionized water (D.I.) was used, to add alkalinity Name Chemical Form Concentration (g/L) Ammonium Chloride NH 4 Cl 107 Ammonium Heptamolybdate (NH 4 ) 6 Mo 7 O 24 H 2 O 0.30 Boric Acid H 3 BO 3 0.02 Calcium Chloride Dihydrate CaCl 2 H 2 O 14.0 Cobalt Chloride Hexahydrate CoCl 2 H 2 O 0.02 Cupric Sulfate CuSO 4 H 2 O 0.50 EDTA EDTA 18.0 Iron Sulfate FeSO 4 H 2 O 1.50 Magnesium Sulfate MgSO 4 H 2 O 90.0 Manganese Chloride MnCl 2 H 2 O 1.50 Potassium Chloride KCl 36.0 Potassium Iodide KI 0.003 Sodium Acetate C 2 H 3 O 2 NaH 2 O 850 Sodium Bicarbonate 1 NaHCO 3 0.168 Sodium Phosphate Dihydrate NaH 2 PO 4 H 2 O 75.5 Sodium Thiosulfate Pentahydrate 2 Na 2 H 3 S 2 H 2 O 0.10 Yeast Extract Yeast Extract 1.00 Zinc Sulfate ZnSO 4 H 2 O 0.70 2 Added when tap water was used, to consume residual chlorine The quality of the synthetic wastewater was dictated by the addition of the chemicals listed in Table 3, with sodium acetate as the main source of COD and sodium bicarbonate to set the alkalinity. Characterization of the synthetic wastewater before being mixed with the sludge in the reactors is presented in Table 4. 16

PAGE 27

17 Table 4. Quality of Synthetic Wastewater Parameter Value Units Alkalinity 100 mg/L CaCO 3 COD 400 mg/L O 2 DO ~ 8.3 mg/L O 2 NH 4 + -N 28 mg/L NH 4 + -N pH ~ 7.6 N/A The SBR operational parameters are listed in Table 5 and a digital image of the entire system is shown in Figure 5. Two thirds of the reactor volume corresponded to synthetic wastewater fed at the beginning of every cycle and decanted at the end of each cycle. Three treatment cycles were operated per day. Therefore, each reactor was fed daily a total of 6 L of synthetic wastewate. Because the reactor volume was 3 L, and the HRT is defined as volume of the reactor di vided by the flow or volume introduced, the set HRT was 0.5 days. This HRT is similar to common values for WWTPs (Metcalf & Eddy 2003). The cycles were automatically operated with a Chrontrol XT-4 (ChronTrol Corporation, San Diego, CA), that switched the feed pump (Masterflex L/S Pump Drive, Model 7518-10, Cole-Parmer Instrument Company, Vernon Hills, IL), waste pump (Masterflex L/S Fixed Flow Drive, Model 7531-01, Cole-Parmer Instrument Company, Vernon Hills, IL), and air supply system. As listed in Table 5, each cycle of 8 hours had a filling period of 10 minutes started with the a 7 hours react cycle. Settling and decanting periods of 45 and 15 minutes, respectively, completed the cycle. Table 5. SBR Operational Parameters Parameter Specification MLSS Volume 3 Liters Volume of Synthetic Wastewater Fed per Cycle 2 Liters Volume of Sludge Seeded 1 Liter HRT 0.5 days Cycles/day 3 Filling 10 min. React/Aeration 7 hours Settling 45 min. Decanting 15 min. Sampling Last 15 min. of react cycle Temperature ~20C (Room temperature) Aeration rate 5 L/min.

PAGE 28

Feed Tanks Pumps Controller Feed Pump Meters CO 2 Controller Electrodes Experimental Reacto r Control Reactor Air Pumps DO mete r Waste Pump Waste Tan k Figure 5. Activated Sludge SBR System for Nitrification with pCO 2 Control 5.2 Experimental Design A description the variables used for the experiments is summarized in Table 6. A total of 10 experiments were designed and conducted for this research. Failure of equipment, water quality, and weather conditions thwarted the operation of experiments 18

PAGE 29

19 1 through 7. Experimental results led to different operation set-ups in later experiments. The system configuration previously described was used for all of the experiments, although several variables including CO 2 supply, SRTs, and water source for synthetic wastewater varied among experiments, as described in Table 6. Table 6. Experimental Design CO 2 Supply 1 Experiment SRT Days Tested per SRT Total Days Tested Period Source Water Source 1 Infinite N/A <1 Continuously Gas cylinder Tap water 2 2 Infinite N/A <1 Continuously Gas cylinder Tap water 3 Infinite N/A <1 Continuously Gas cylinder Tap water 4 8-day 4-day 8 3 11 Entire react cycle Chamber Tap water 5 Doubled 8-day 4-day 16 1 17 Entire react cycle Chamber Tap water 6 8-day 6 6 Entire react cycle Chamber Tap water 7 8-day 6-day 4-day 2-day 8 6 4 1 19 Entire react cycle Chamber D.I. Water 8 8-day 6-day 8 3 11 Entire react cycle Chamber Bottled D.I. Water 9 8-day 6-day 4-day 2-day 8 6 4 2 20 Last 5 hours of the react cycle Chamber Bottled D.I. Water 10 Tripled 8-day 6-day 4-day 2-day 24 18 12 6 60 Last 5 hours of the react cycle Chamber R.O. Water 1 Carbon dioxide (1%) to the experimental reactor 2 No acetate added 5.2.1 Solids Retention Time Experiments 1 through 3 were designed for an infinite SRT (no wasting of biomass) to rapidly determine the impact of CO 2 on nitrification rates. Once promising results were obtained, Experiments 4 and 6 were set with an 8-day SRT, sequentially

PAGE 30

20 dropped to 4 days to evaluate the impact of CO 2 on nitrification during a short SRT. Similarly, Experiment 5 was designed for consecutive 8 and 4-day SRTs but each SRT was to be maintained twice (i.e. 8-day-SRT: 16 days) to observe the effects of CO 2 under more stable conditions. To avoid a drastic change in the SRT, Experiments 7 through 10 were designed to drop the SRT from 8 days sequentially to 6, 4, and 2 days. Finally, Experiment 10 was designed to maintain 3 times each SRT to evaluate the impact of CO 2 on nitrification under steady-state conditions. 5.2.2 Supply of Air and CO 2 to the Experimental Reactor Air was supplied to both reactors at a rate of 1 L/min during Experiments 1 through 4. However, a DO concentration as low as 0.1 mg O 2 /L was observed in both reactors during the first hour of the cycle when acetate was being consumed. As a consequence, Experiment 4 (in which a finite SRT was used) presented proliferation of filamentous bacteria favored by low DO levels. Hence, poor settling performance was observed after 6 days, which resulted in bulking and washout of biomass. Therefore, for subsequent experiments, the aeration rate was set to the maximum (5 L/min), and DO concentrations were maintained above 3 mg O 2 /L during the entire react cycle. Carbon dioxide (1%) was supplied directly from a 1% CO 2 gas cylinder during Experiments 1 through 3, but this approach was inconvenient due to operational costs. Therefore, later experiments used the CO 2 chamber shown in Figure 3(B). Experiments 4 through 8 were conducted by supplying CO 2 during the entire react cycle (7 hours), whereas for Experiments 9 and 10 CO 2 was added during the last 5 hours of the react cycle (Table 6). The effects of the different CO 2 supply strategies are described in the results and discussion section. 5.2.3 Water The synthetic wastewater fed in all the experiments was prepared as listed in Table 3. Experiments 1 through 6 were conducted using tap water to prepare the feed. Due to interferences with the equipment caused by the ionic strength of the tap water, deionized (DI) water provided by the Biological Sciences Lab (USF; Tampa, FL) was

PAGE 31

21 used for Experiment 7. Contamination of the deionization equipment required the replacement of this DI water for bottled DI water (Culligan Water. Water from Floridas municipal county wells; processed by reverse osmosis, deionization, and ozonation; DT Water, Ft. Myers, FL), which was used for Experiments 8 and 9. Due to availability of a new reverse osmosis (RO) unit in the Kopp Engineering Building, Experiment 10 was conducted with this RO water. The water used for Experiment 10 had a conductivity of 6 S, measured with an YSI 35 conductivity meter (YSI Inc., Yellow Springs, OH). 5.3 Data Collection and Sample Analyses The strategy for data and sample collection is summarized in Table 7. Data and samples to determine NH 4 + and NO 3 formation rates, pH, and DO were collected daily during the first entire react cycle (7 hours). Samples of MLSS for settling evaluation and biomass analyses were collected daily by the end of the third react cycle. Ammonium and NO 3 concentrations expressed as nitrogen (NH 4 + -N and NO 3 -N respectively) were measured every 30 minutes during the react cycle to determine nitrification rates using ion selective electrodes (Ammonium combination glass body electrode, Cole-Parmer 27502-03 and Nitrat e combination glass body electrode, ColeParmer 27502-31, Cole-Parmer Instrument Company, Vernon Hills, IL) and ion meters (Oakton Benchtop Ion 510 Meter and Oakton Ion 6 Meters, Cole-Parmer Instrument Company, Vernon Hills, IL). The ion selective electrodes were calibrated daily before the first react cycle started. The ammonium electrode used a 0.1M NaCl filling solution (Cole Parmer 27503-78 reference filling solution, Cole-Parmer Instrument Company, Vernon Hills, IL) and was calibrated with a 1000 ppm NH 4 + -N Ammonium standard solution (prepared in the laboratory with reagent-grade NH 4 Cl) and a 5M NaCl Ionic Strength Adjuster (ISA) prepared in the laboratory. The nitrate electrode used a 0.1M (NH 4 ) 2 SO 4 filling solution (Cole Parmer 27503-79 reference filling solution, ColeParmer Instrument Company, Vernon Hills, IL) and was calibrated with a 1000 ppm NO 3 -N nitrate standard solution (prepared in the laboratory with reagent-grade NaNO 3 ) and a 1M NaSO 4 Ionic Strength Adjuster (ISA) prepar ed in the laboratory. Appendix 1 describes the calibration procedure in detail.

PAGE 32

Samples for TSS, volatile suspended solids (VSS), and COD analysis were collected once per day from the mixed liquor during the last 15 minutes of the react cycle, to evaluate the system performance. A preliminary test to evaluate the settling performance was conducted daily. For this test, 100 mL of MLSS were withdrawn from each reactor and settle for 30 minutes in a 100 mL graduate cylinder. The volume of solids settled, reported as sludge blanket in mL, was collected daily as an indication of settling performance. A large sludge blanket indicated a poor settling performance. Sludge blankets greater than 40 mL usually indicated poor settling and biomass washout. Figure 6 shows an example of the sludge blanket volume measured after settling for 30 minutes. Figure 6. Sludge Blanket Volume for Settling Performance Preliminary Test. The sludge volume index (SVI) was calculated by Equation 4 below. Values of SVI greater than 150 mL/g indicate poor settling performance and abundance of filamentous bacteria (Metcalf & Eddy 2003). Sludge Blanket Volume g mg 1000 (mg/L) TSS L 0.1(mL) blanket sludge (mL/g) SVI (4) 22

PAGE 33

23 Table 7. Data and Sample Collection Strategy Parameter Frequency Purpose NH 4 + -N (mg/L) Every 30 minutes Ammonia oxidation rates NO 3 -N (mg/L) Every 30 minutes Nitrate formation rates COD Daily Treatment efficiency Solids Daily Efficiency/performance: TSS, VSS, SVI pH Twice per cycle Operation control DO (mg/L O 2 ) Beginning of cycle Operation control Biomass: DNA Daily Bacteria Identification/cloning Biomass: RNA Daily Bacteria growth response Biomass: FISH Daily Bacteria classification/enumeration Dissolved oxygen (Traceable Digital Oxygen Meter, Control Company, Friendswood, TX) and pH (Waterproof pHTest r 3+ double Junction, Oakton Instruments, Vernon Hills, IL) were measured periodically. 5.3.1 Analytical Methods A digital image of the apparatus used to filter the samples for solids analyses is shown in Figure 6. Samples of 45 mL from the MLSS were collected for the analysis of SS in 50 mL conical tubes and kept refrigerated at 4C. The apparatus to filter the samples was built as specified by the Standard Methods for the Examination of Water and Wastewater Analysis (AWWA, 2000), section 2540C.2a.2c. This apparatus consisted of 6 Pyrex TM filter flasks (Fisher Scientific., Pittsburgh, PA), each connected through a vacuum manifold to a Leybold Trivac D8B vacuum pump (Leybold-Heraeus Vacuum Products, Inc., Export, PA). A rubber stopper inserted in the neck of each flask made it suitable to hold one Coors TM porcelain Gooch filtering crucible (Fisher Scientific., Pittsburgh, PA). Whatman glass microfibre filters (934-AH TM Whatman Inc., Clifton, NJ) were inserted in the crucibles for filtration of the samples.

PAGE 34

Figure 7. Filtration Apparatus for SS Analyses. The procedures for the preparation and determination of TSS and VSS were followed as described in the Standard Methods for the Examination of Water and Wastewater Analysis (AWWA, 2000) sections 2540D and 2540E respectively. Samples were dried for 1 hour at 103 105C using a Fisher Isotemp TM 516G lab oven (Fisher Scientific, Pittburgh, PA), and then kept in a Nalgene TM vacuum dessicator (Nalgene, Rochester, NY) for cooling down and balanci ng the temperature before being weighed. Dried samples were weighed using an APX-402 balance (Denver Instrument Company, Arvada, CO). The drying, cooling, and wei ghing steps were repeated (usually twice) until values stabilized or the weight change was less than 4% or 0.5 mg; whichever was less. Mass values were used to calculate the TSS concentration as indicated by Equation 5. For the determination of VSS, the weighed samples were ignited in a F48015 Thermolyne Furnace (Barnstead International/Electrothermal, Essex, United Kingdom) at 550C for 15 minutes. The ignited samples were repeatedly cooled in the dessicator and weighed, until the values stabilized. The same criteria used for the TSS concentrations were used to determine the final weight of the sample. Equation 6, below, was used to calculate the VSS, according to the standard methods. Triplicate samples were analyzed for both TSS and VSS determination. 24

PAGE 35

mL volume, sample mL/L1000mg B1-A (mg/L) TSS (5) mL volume, sample mL/L1000mg B2-A (mg/L) VSS (6) Where: A = weight of filter and crucible + dried residue, mg, B1 = weight of filter and crucible, mg, and, B2 = weight of residue + filter and crucible after ignition, mg. Sample volume = 10 mL Samples for COD analysis were withdrawn from both reactors (10 mL of MLSS) and settled for 30 minutes. Successively, the supernatant was filtered with Fisherbrand TM 25 mm syringe filters (Fisher Scientific., Pittsburgh, PA). Filtered samples were preserved in 15 mL conical tubes at -20C. Later determination of COD was performed using the Reactor Digestion Method 8000 (Jir ka and Carter 1975) for the range 3 150 mg/L, approved by the United States Environmental Protection Agency (USEPA) for wastewater analyses 1 The vials used for this procedure (Digestion solution for COD 0150 ppm range, HACH Company, Loveland, CO) were mixed with 2 mL of sample as indicated in the Method 8000 and digested for 2 hours at 150C in a digital reactor block DRB200 (HACH Company, Loveland, CO). Vials were placed in a rack for cooling down to room temperature (~21C). A portable spectrophotometer DR/2400 (HACH Company, Loveland, CO) adjusted to a wavelength of 420 nm (program 430 COD LR) as indicated by the Method 8000 was used to read the COD concentrations of the samples. A vial mixed with 2 mL of DI water was used as a blank and additional vials each mixed with 300 mg/L COD standard solution at different dilutions were digested to check the calibration curve of the spectrophotometer with known concentrations of COD. 25 1 Federal Register, April 21, 1980, 45(78), 26811-26812

PAGE 36

6. Model of the Impact of pCO 2 on pH Prior to experimental work a model was developed to predict the effect of CO 2 on the pH for a range of typical buffer capacities (alka linities). Because the alkalinity was adjusted with NaHCO 3 (Table 3), a charge balance of the i ons in solution contributing to alkalinity was used as shown in Equation 7. Other species were considered negligible. ]OH[][CO2]HCO[]Na[]H[-2 3 3 (7) Equation 7 was rearranged to make the pH a function of [Na + ] and pCO 2 by using Equations 8 to 11: ]CO[H ]HCO][H[ 10:K* 32 3 35.6 a,1 (8) ]H[ ]CO[HK ]HCO[* 32a,1 3 (9) ][HCO ]CO][H[ 10:K3 2 3 33.10 a,2 (10) ]H[ ][HCOK ]CO[3 a,2 2 3 (11) Assuming that the atmospheric CO 2 is in equilibrium with CO 2(aq) and defining [H 2 CO 3 ] as a function of pCO 2 Equation 12 resulted (Snoeyink and Jenkins, 1980): 26

PAGE 37

2 5.1 2 H 2(aq) 32pCO10pCOK CO]COH[ (12) where, K H = 10 -1.5 Henrys constant Equation 13 below was derived by substituting Equation 12 into Equation 9: ]H[ pCOKK ]H[ ]CO[HK ]HCO[2Ha,1 32a,1 3 (13) By using the water constant (Equation 14), [OH ] was expressed in terms of [H + ] in Equation 15: ]OH][H[10K14 w (14) ]H[ K ]OH[w (15) Since the pH of pure water is less than 9, [CO 3 2] in Equation 7 is considered negligible. Substituting Equations 13 and 15 into Equation 7 resulted in Equation 16: ][H K ]H[ pCOKK ]Na[]H[w 2Ha,1 (16) Equation 17 below shows the pH as a function of pCO 2 : 2 )KpCOKK(4]Na[]Na[ ]H[w2Ha,1 2 (17) Values of pH were calculated using Equation 17 for different values of pCO 2 and concentrations of NaHCO 3 In this research, alkalinity was adjusted by the addition of NaHCO 3 Therefore, to model the impact of CO 2 on the wastewater pH, Na + concentrations correspondent to typical alkalinity values in the wastewater were used. Table 8 presents the information used to compute the Na + concentrations correspondent to three different alkalinity values (as mg/L of CaCO 3 ). 27

PAGE 38

Table 8. Molecular Weight and Concentration of CaCO 3 and NaHCO 3 in eq/mole and mg/meq Compound MW eq/mole g/eq CaCO 3 100 g/mole 2 50 g CaCO 3 /eq NaHCO 3 84 g/mole 1 84 g NaHCO 3 /eq To express the compounds in grams of the compound per equivalent, as reported in Table 8, Equation 18 was used: eq Compund g eq/mole# (g/mole) MW (18) Table 9 presents three alkalinity values typical of wastewater, expressed as CaCO 3 and their correspondent Na + concentrations, when alkalinity was adjusted with NaHCO 3 Table 9. Typical Alkalinity Values in Wastewater and their Correspondent Concentrations of Na + -when Adjusting Alkalinity with NaHCO 3 Alkalinity (mg/L as CaCO 3 ) Alkalinity (meq/L) NaHCO 3 (g/L) NaHCO 3 (M) 50 1 0.084 0.001 100 2 0.168 0.002 200 4 0.336 0.004 Values of alkalinity in meq/L shown in Table 9 were computed using Equation 19. The Na + concentrations were computed using Equation 20 (g Na + /L) and Equation 21 (M of Na + ). /meqCaCO mg 50 )CaCO as ALK(mg/L )(meq/L ALK 3 3 (19) mg 1000 g 1 meq NaHCO mg 84) ALK(meq/L )(g/L NaHCO3 3 (20) 3 3 3NaHCO ofMW (g/L) NaHCO )(M NaHCO (21) 28

PAGE 39

Figure 8 illustrates the pH obtained for three typical alkalinity values of wastewater. Alkalinity was adjusted by adding NaHCO 3 to the wastewater. The concentrations of the Na + ions in solution are shown in Figure 8 representing different alkalinity values. For instance, a concentration of 0.001 M Na + corresponds to an alkalinity of 50 mg/L as CaCO 3 a concentration of 0.002 M Na + corresponds to an alkalinity of 100 mg/L as CaCO 3 and a concentration of 0.004 M Na + corresponds to an alkalinity of 200 mg/L as CaCO 3 (Table 9). 5.0 6.0 7.0 8.0 9.0 10.0 1.E-04 1.E-03 1.E-02 1.E-01 pCO2pH 0.001 M 0.002 M 0.004 M [ Na + ] 0.03% CO2 Control R eactor 1% CO2Experimental Rector Figure 8. pH as a Function of pCO 2 and Alkalinity An increase in pCO 2 was found to cause a decrease in pH and for any given alkalinity. The model predicted a difference between the control and the experimental reactors of approximately 1.5 pH units. For the alkalinity of the synthetic wastewater used in the experiment (100 mg/L as CaCO 3 ) the pH was predicted to be approximately 8.5 units in the control reactor and 7.2 units in the experimental reactor. The concentrations of interest, 0.03% CO 2 for the control and 1% CO 2 for the experimental reactors are circled in Figure 8. Both pH values are within the pH range for typical activated sludge systems. For all experimental work, 1% CO 2 was used for the experimental reactor. 29

PAGE 40

30 7. Results and Discussion 7.1 Preliminary Results Due to equipment failure, adverse weather (hurricane season), and contamination of the distilled water used for synthetic wastewater, the first 7 experiments were conducted without much success. After the seventh experiment, bottled deionized water and a local source of RO water were used to prevent contamination. Three subsequent experiments were conducted successfully (Experiments 8, 9, and 10 in Table 6). 7.2 Experiment 8: Supply of CO 2 during the Full React Cycle Using the experiment set-up and operational variables described in Table 6, the first significant experiment was conducted for a complete 8-day-SRT, and for 3 more days with an SRT of 6 days. The experiment was terminated due to washout of biomass caused by poor settling in the experimental reactor. Ammonium oxidation rates were inconclusive due to malfunctioning of the NH 4 + electrode. 7.2.1 Nitrate Formation Rates Exp. 8 Figure 9 presents NO 3 -N concentrations as a function of time for data collected in both reactors during the first cycle of day 3 (E xperiment 8). The slope of the trend line equation indicates the nitrate formation rate. As shown in Figure 9, the NO 3 formation rate in the experimental reactor (0.014 mg NO 3 -N/L min) during day 3 was approximately 5 times higher than that of the control reactor (0.003 mg NO 3 -N/L min). This trend was observed during all the experiments.

PAGE 41

y = 0.003x + 0.2904 R2 = 0.9864 y = 0.014x + 0.616 R2 = 0.999 0.0 1.0 2.0 3.0 4.0 5.0 6.0 050100150200250300350 Time (minutes)NO3 --N (mg/L) Control Experiment Figure 9. Nitrate Formation Day 3, Experiment 8 Figure 10 shows daily nitrification rates obtained for both reactors throughout the experiment, as described above. Nitrate formation rates in the control reactor during the first 2 days of the experiment were significantly low due to the new conditions for the biomass. Rates increased rapidly up to day 3 with a subsequent slight increase until day 7, when a maximum rate of 0.004 mg NO 3 -N/L-min was reached. A decrease in nitrification rates was observed after day 8 as a consequence of the reduction of the SRT from 8 to 6 days. The control reactor recove red nitrification after 3 of the reduction of the SRT (day 11) indicating a positive response of nitrite oxidizing bacteria to low SRTs. For the entire experiment, the average nitrate formation rate in the control reactor was 0.002 mg NO 3 -N/L-min. However, the experiment was not continued further due to washout of biomass and loss of nitrif ication in the experimental reactor. 31

PAGE 42

0.000 0.002 0.004 0.006 0.008 0.010 0.012 0.014 0123456789101112 Time (days)NO3 --N (mg/L-min ) 32 Contro l Experi ment 8-day SRT 6-day SRT Figure 10. Nitrate Formation Rates Experiment 8 Similarly, the experimental reactor showed a low nitrate formation rate during the first day of the experiment with an average rate of 0.008 mg NO 3 -N/L-min. However, a rapid increase was observed until day 3 when a maximum rate of 0.014 mg NO 3 -N/Lmin was reached. A constant decrease in the nitrate formation rates was observed until day 8 when the SRT was lowered to 6 days. From day 8, the rates decreased rapidly down to 0.0004 mg NO 3 -N/L-min (day 11). Unintentional lost of biomass due to washout was observed after the reduction of the SRT on day 8 (~ 500 mL MLSS/cycle), day 9 (300 mL MLSS/cycle), and day 11 (1000 mL MLSS/cycle), which had a significant impact on nitrate formation rates and affected the overall performance of the process. Both reactors showed poor nitrate formation rates during the first day of the experiment due to the reduction of the SRT from the original WWTP (22-day-SRT) to the conditions set for the experiment. However, a positive effect on the nitrate formation rates was observed in the experimental reactor, which showed rates 3 times higher in average and up to 5 times higher than the control. In spite of this, a significant decrease on nitrate formation rates was observed in the experimental reactor after day 7 when 100 mL of MLSS was lost due to poor settling. Hence, the reduction in nitrification rates was

PAGE 43

33 related to poor settling and washout of biomass in the experimental reactor, which as opposite from the control reactor, did not recover from the reduction of the SRT. 7.2.2 Settling Performance Exp. 8 Figure 11 illustrates the settling performance for both reactors during Experiment 8. The settling performance was evaluated by pouring 100 mL of MLSS in a graduated cylinder to settle for 30 minutes, as described in the materials and methods section. The volume of solids settled was reported as sludge blanket in mL. Sludge blanket volumes above 40 mL/100 mL were related to poor settling and potential washout of biomass. The control reactor showed a stable settling performance with a slight increase on day 8. The maximum level reached was 29 mL/100mL in the 30-minute settling test. No loss of biomass was observed in the control reactor and a normal settling performance was maintained during the course of the experiment. In contrast, poor settling in the experimental reactor resulted after a few days of operation, with the sludge blanket increasing over time to more than 65 mL per every 100 mL (day 8) and up to 90 mL the last day of the experiment (day 11). Considering that two thirds of the reactor volume were wit hdrawn every cycle, and due to poor settling only one third of the volume corresponded to clarified water, washout of biomass was evident in the experimental reactor. For in stance, after day 7 an average of 100 mL of MLSS per cycle was unintentionally wasted due to poor settling, accounting for loss of 3% of the biomass every cycle. The experi mental reactor did not recover from biomass washout and the experiment was ended due to loss of nitrification.

PAGE 44

0 20 40 60 80 100 0123456789101112 Time (days)Sludge Blanket (m L Control Experiment 8-day SRT 6-day SRT Figure 11. Sludge Blanket Volume per 100 mL of Sample Experiment 8 By comparing both reactors, the control reactor showed a normal response to the reduction of the SRT and adequate settling performance, whereas the experimental reactor had a poor settling performance. Poor settling was not likely to have been associated with low DO levels for it was maintained above 3 mg O 2 /L as described in the experimental design section. Settling in the experimental reactor, however, seemed to have been affected by the supply CO 2 during the entire react period. These results suggest that high concentrations of CO 2 during the first 2 hours of aeration despite DO levels were greater than 3.0 mg/L O 2 may have favored the growth of a particular facultative type of filamentous bacteria able to synthesize both organic and inorganic carbon. Such indication is consistent with recent research of Thiothrix spp. which has been proved to be a very versatile facultative heterotrophic organism with mixotrophic and chemolithoautotrophic potentia l. This filamentous bacteria was demonstrated to fix bicarbonate into cell biomass while in the presence of acetate (Nielsen et al. 2000). Additional ability of th is type of filamentous bacteria to have certain activity in presence of nitrate during anaerobic periods, might have given it a selective advantage over other commonly-found filamentous bacteria in the activated sludge systems. Therefore, it is suggested that the supply of CO 2 during the period when 34

PAGE 45

acetate is available, may have selected for heterotrophic filamentous bacteria that are also facultative autotrophs. 7.2.3 COD Removal Exp. 8 Figure 12 shows the percentage removal of COD throughout the experiment. Synthetic wastewater used to feed the react ors had a COD concentration of 400 mg/L as O 2 was. However, since only two thirds of the reactor volume was fed every cycle (for the other third corresponded to the seeded activated sludge), two thirds of the synthetic concentration (267 mg/L O 2 ) was used as the initial COD concentration for the calculation of removal efficiencies. 90% 92% 94% 96% 98% 100% 0246810 Time (days)COD Remova 12 l Control Experiment 8-day SRT 6-day SRT Figure 12. COD Removal Efficiencies Experiment 8 Both reactors presented removal efficiencies greater than 96% corresponding to a supernatant concentration of 7.0 and 6.0 mg/L as O 2 in the control and experimental reactors respectively. Average removal of COD was 96% for the control reactor (9.5 mg/L as O 2 ) with a standard deviation of 0.8%, and 97% for the experimental reactor (7.5 mg/L as O 2 ) with a standard deviation of 0.7%. According to these values, no significant difference was observed in COD removal by the addition of CO 2 The minimum removal 35

PAGE 46

36 efficiency required of secondary treatment was met (90%) and the concentrations in the supernatant were always below 30 mg/L O 2 as required by the current regulations. 7.2.4 Summary of Results Exp. 8 A positive impact of adding 1% CO 2 during aeration was evident during Experiment 8, for nitrate formation rates in the experimental reactor were up to 5 times higher than the control. Maximum nitrate formation rates for each reactor were 0.004 and 0.014 mg NO 3 -N/L-min (control and experiment respectively), and average rates were 0.002 and 0.008 mg NO 3 -N/L-min for the control and experimental reactors respectively. However, results suggested an adverse impact of CO 2 on settling performance, when supplied during the entire react period. Sludge blanket volumes above 40 mL/100mL and washout of biomass we re observed in the experimental reactor whereas the control reactor showed an adequate settling performance. Finally, no impact of CO 2 was evidenced on COD removal efficiencies, which was greater than 90% in both reactors. 7.3 Experiment 9: Supply of CO 2 during the Last 5 Hours of the React Cycle Based on findings in Experiment 8, CO 2 was supplied to the experimental reactor after the first 2 hours of every cycle when all the acetate had been consumed by the heterotrophic bacteria and the DO levels were maintained above 3 mg/L. Measurement of NO 3 -N concentrations were collected every 30 minutes, while concentrations of NH 4 + could not be measured due to electrode failu re. Experiment 9 was conducted for a period of 20 days lowering the SRT consecutively from 8 days to 6, 4, and 2 days. 7.3.1 Nitrate Formation Rates Exp. 9 Nitrate formation rates were calculated daily as described in Experiment 8 and are presented in Figure 13. Nitrate formation rates in the control reactor dropped fast the first 3 days of the experiment and then increased until day 6 when a maximum rate of 0.004 mg NO 3 -N/L-min was reached. Similar to the results obtained in Experiment 8, a steady performance was observed from day 6 through day 10. From day 10 through the

PAGE 47

end of the experiment (day 20) the nitrate formation rates declined constantly due to poor settling of the solids which occurred since the SRT was lowered. The average nitrate formation rate obtained in the control reactor was 0.002 mg NO 3 -N/L-min, identical to that obtained during Experiment 8. 0.000 0.005 0.010 0.015 0.020 01234567891011121314151617181920 Time (days)NO3 --N (mg/L-min ) Control Experiment ` 8-day SRT 6-day SRT 4-day SRT 2-day SRT Figure 13. Nitrate Formation Rates Experiment 9. In the experimental reactor, nitrate form ation rates rapidly increased from day 2 to 6 when the maximum rate was reached (0.016 mg NO 3 -N/L-min). This maximum rate was slightly higher than the maximum rate obtained during Experiment 8 (0.014 mg NO 3 -N/L-min). A drastic decrease on nitrate formation rates was observed from day 10 to the end of the experiment. Even though on day 13 nitrate formation rates recovered from the reduction of the SRT, on day 16 the SRT was dropped again to 4 days causing another decline of nitrate formation rates. An average rate of 0.008 mg NO 3 -N/L-min was calculated for the experimental reactor. Average, maximum, and minimum rates for the experimental reactor during Experiment 9 were close but slightly higher than those from Experiment 8, thereby indicating a positive response to the addition of CO 2 A comparison of both reactors shows that the nitrate formation rates in the experimental reactor were consistently higher, with a maximum ratio of 6 37

PAGE 48

38 (experiment/control) during day 3. The average ratio of nitrate formation rates throughout the experiment was 3 (experiment/control). Although nitrification rates decreased as the SRT was dropped to 6, 4, and 2 days (days 8, 14, and 18 respectively) as shown in Figure 13, the average ratio was maintained. This suggests a positive effect of CO 2 in nitrate formation rates comparable to the results obtained for Experiment 8. 7.3.2 Settling Performance Exp. 9 Figure 14 shows the sludge blanket volume that was measured daily for both reactors. An average sludge blanket of 32 mL was registered throughout the experiment in the control reactor. Settling performance in this reactor was normal during the first SRT although a rapid increase in the sludge blanket volume was observed after day 8 when the SRT was lowered to 6 days. As a consequence of the reduction of the SRT, poor settling performance was observed with a maximum sludge blanket volume of 66 mL on day 10, and consequent washout of bi omass. Loss of biomass due to poor settling was observed for sludge blanket volumes above 40 mL. From days 10 through 16 an average value of 40 mL was registered, with a new increase on day 17 (52 mL) which again resulted in loss of biomass. It was observed consistently that 2 or 3 days after lowering each SRT (days 10 and 17) the sludge blanket volume peaked, suggesting that the poor settling performance was related to the reduction of the SRT.

PAGE 49

39 0 10 20 30 40 50 60 70 01234567891011121314151617181920 Time (days)Sludge Blanket (m l Control Experim ent 8-day SRT 6-day SRT 4-day SRT 2-day SRT Figure 14. Sludge Blanket Volume per 100 mL of Sample Experiment 9 In contrast, the experimental reactor experienced excellent settling performance with an average sludge blanket of 25 mL throughout the experiment. The maximum sludge blanket of 33 mL (day 10), indicated normal operation of the experimental reactor without loss of biomass. Compared to the results obtained in Experiment 8, an improvement in settling performance was obtained by supplying CO 2 after 2 hours from the beginning of the react cycle. These results suggest that the conditions given in Experiment 8 could have selected for f acultative chemolithoautotrophic filamentous bacteria able to synthesize inorganic carbon in presence of acetate, and are consistent with recent research in the field found else where (Odintsova et al. 1993; Tandoi et al. 1994; Nielsen et al. 2000). Moreover, these re sults are consistent with recent findings that report inorganic carbon as an effective remedy to poor sludge settling in BNR systems (Wett and Rauch 2003). However, it is presumed that when calcium bicarbonate is used as the source of inorganic carbon, di ssociation of the ions in the water could increase the availability of the Ca +2 ions which act as a coagulants and favor floc formation. Therefore, future research should be addressed to identify the real effect of inorganic carbon (e.g. CO 2 ) on settling performance.

PAGE 50

A comparison of both reactors throughout the experiment, show the experimental reactor maintaining a stable settling performance while the control reactor was seriously affected by the reduction of the SRT on days 8, 14, and 18. Small sludge blanket volumes observed after day 18 (< 20 mL/100 mL) were due to the reduction of the SS concentration. From day 0 through 8, settling for both reactors was very similar until the SRT was lowered, a change that impacted the control reactor significantly but not the experimental reactor. These results compared to the settling performance in both reactors during Experiment 8, suggest a positive effect of CO 2 when supplied after the organic carbon has been consumed and while a high DO concentration is maintained. 7.3.3 COD Removal Exp. 9 Figure 15 shows the percentage removal of COD throughout the experiment. COD concentrations in the supernatant and removal efficiencies were similar in both reactors, and were also consistent with results obtained during Experiment 8. 90% 92% 94% 96% 98% 100% 02468101214161820 Time (days)COD Remova l Control Experiment 8-day SRT 6-day SRT 4-day SRT 2-day SRT Figure 15. COD Removal Efficiencies Experiment 8 Both reactors had a minimum removal efficiency of 96% and an average COD removal of 97% corresponding to 9.0 mg/L as O 2 in the control reactor with a standard 40

PAGE 51

41 deviation of 0.3%; and 7.8 mg/L as O 2 with a standard deviation of 0.8% in the experimental reactor. As found in Expe riment 8, no significant impact of CO 2 was observed in the COD removal performance. 7.3.4 Summary of Results Exp. 9 The results of Experiment 9 indicate that for high concentrations of CO 2 nitrifying bacteria grow faster and are able to maintain nitrification at lower SRTs without affecting the performance of the system (i.e. settling and COD removal). Maximum nitrate formation rates for each reactor were 0.004 and 0.016 mg NO 3 -N/Lmin (control and experimental respectively), and average rates were 0.002 and 0.007 mg NO 3 -N/L-min for the control and experimental reactors respectively. Peak sludge blanket volumes above 40 mL were observed twice in the control reactor whereas the experimental reactor showed an adequate settling performance ( 33 mL). COD removal efficiencies greater than 90% throughout th e experiment in both reactors indicate no correlation between COD removal and CO 2 concentration. Similarity in the results for Experiments 8 and 9, indicating a positive effect of CO 2 on nitrate formation rates strengthen hypotheses 1 and 2 of this research. Differences in settling performance in the expe rimental reactor in Experiments 8 and 9, suggest a negative effect of supplying CO 2 while acetate is still readily available for facultative filamentous bacteria. These preliminary results will lead to future research to evaluate the settling performance as a function of CO 2 concentrations, aeration periods, and SRT. 7.4 Experiment 10: Tripled SRT Confirmatory Experiment To confirm results from Experiment 9, Experiment 10 was designed with the same operational parameters although each SRT was maintained 3 times (Table 6) to evaluate the efficacy of the treatment under steady-state conditions. Solids concentrations (TSS and VSS) were determined to calculate the SVI (Equation 4) and the specific nitrate formation rates. Specific Nitrate formation rates were normalized to the VSS concentration by using Equation 22, and expressed as mg NO 3 -N/g VSS-min.

PAGE 52

gmg 1000 (mg/L) VSS min)N/L-NO (mg rate formation NO minVSS g N-NO mg3 3 3 (22) 7.4.1 Nitrate Formation Rates Exp. 10 Nitrate formation rates shown in Figure 16 were calculated as in previous Experiments and expressed as mg NO 3 -N/L-min. The control reactor showed had a maximum rate of 0.005 mg NO 3 -N/L-min (day 0) and an average rate of 0.001 mg NO 3 N/L-min. A minimum nitrate formation rate of 0.0003 mg NO 3 -N/L-min was observed on day 22. Nitrate formation rates were low throughout the experiment with rates slightly higher during the 8-day-SRT period (d ays 0 through 24). This period showed an average rate of 0.0019 mg NO 3 -N/L-min whereas the 6, 4, and 2-day SRT periods averaged 0.0009, 0.0007, and 0.0008 mg NO 3 -N/L-min respectively. A slight increase on nitrate formation rates was observed from days 28 through 32, towards the end of the first 6-day-SRT period. A decrease was again observed from day 32 until the end of this period (day 42). During the 4 and 2-day SRT periods (days 42 through 60), the nitrate formation rates did not increase significantly and were steadily at low values. 0.000 0.005 0.010 0.015 0.020 024681012141618202224262830323436384042444648505254565860 Time (days)NO3 --N (mg/L-min ) Control Experiment 8-day SRT 6-day SRT 4-day SRT 2-day SRT Figure 16. Nitrate Formation Rates Experiment 10 42

PAGE 53

43 The experimental reactor showed an average rate of 0.005 mg NO 3 -N/L-min throughout the experiment, with a maximum of 0.012 mg NO 3 -N/L-min (day 0) and a minimum of 0.0024 (day 1). This variation during the first few days of the experiment was due to the adaptation of the biomass to the new conditions (SRT, CO 2 concentration, and system configuration). However, the highest rates were observed during the 8-daySRT period (days 0 through 24). During this period an average nitrate formation rate of 0.0068 mg NO 3 -N/L-min was observed, which was higher than the 6, 4, and 2-day SRT periods that averaged 0.0037, 0.0044, and 0.0036 mg NO 3 -N/L-min respectively. Fluctuating nitrate formation rates during th e first 8-day SRT were followed by a steady period during the second 8-day SRT (days 9 through 16). A decrease on nitrate formation rates was observed from day 16 through day 32, due to a decrease in settling performance (increasing SVI) observed during the same period. After day 32 nitrate formation rates increased and were steady through the end of the experiment (day 60), which led to similar average values among the 6, 4, and 2-day SRT periods, as indicated before. Therefore, no significant difference was evident among SRTs in the experimental reactor, except for the 8-day-SRT period, which showed rates slightly higher. A comparison of both reactors suggests a positive effect of CO 2 on nitrate formation rates, for the experimental reactor showed higher rates throughout the experiment. Rates in the experimental reactor were 5 times higher in average and up to 12 times higher (day 50) than in the control reactor. Additionally, a faster recovery of nitrate formation was observed in the experimental reactor after day 32 when settling performance recovered, as discussed later. 7.4.2 Total and Volatile Suspended Solids Exp. 10 The concentrations of TSS and VSS were determined as described in the materials and methods section and are shown in Figures 16 and 17 respectively. Figure 19 shows the percentage content of VSS. In the control reactor, both TSS and VSS concentrations did not change significantly during the first 8-day-SRT period (days 0 through 24), although a reduction was observed towards the end (day 22). Average solids concentration of 1,696 mg/L (TSS) and 1,560 mg/L (VSS) were observed during

PAGE 54

44 the 8-day-SRT period. These concentrations corresponded to an average 92% (VSS/TSS) concentration. Both TSS and VS S concentrations were oscillating during the entire 6-day-SRT period (days 25 through 42) when poor settling performance was observed. Reduction of solids was evident on days 28 and 32 through 38 due to washout of biomass on day 27 (~50 mg MLSS/L-cycle) and days 32 through 38 (~100 mg MLSS/L-cycle). Average concentrations were 1,456 mg/L (TSS) and 1,263 mg/L (VSS), with an average VSS content of 87%, slightly lower than the content during the precedent period. This values are consistent with the loss of biomass observed. During the 6-daySRT period, due to the loss of biomass, the concentration of solids decreased. In contrast, the 4-day-SRT period (days 43 th rough 54) did not experience washout of biomass. This is reflected in the observed steady concentrations with just a slight increase towards the end of the period. Average concentrations were 1,350 mg/L (TSS) and 1,150 mg/L (VSS). A 85% of solids were VSS, very similar to the precedent period, indicates that no significant loss of biomass occurred and the system recovered from washout during the 6-day-SRT period. However, the reduction of the SRT to 2 days (day 54), caused a drastic decrease in the solids concentrations. For instance, average TSS and VSS concentrations of 864 mg/L and 791 mg/L were observed during this 2-daySRT period (days 54 through 60). However, a higher biomass content indicated for a 92% (VSS/TSS) was observed.

PAGE 55

500 1,000 1,500 2,000 2,500 024681012141618202224262830323436384042444648505254565860 Time (days)TSS (mg/L) 45 Cont rol Exper iment 8-day SRT 6-day SRT 4-day SRT 2-day SRT Figure 17. Total Suspended Solids Experiment 10 500 1,000 1,500 2,000 024681012141618202224262830323436384042444648505254565860 Time (days)VSS (mg/L) Control Experiment 8-day SRT 6-day SRT 4-day SRT 2-day SRT Figure 18. Volatile Suspended Solids Experiment 10

PAGE 56

50% 60% 70% 80% 90% 100% 024681012141618202224262830323436384042444648505254565860 Time (days)VSS / TSS (%) 46 Contro l Experi ment 8-day SRT 6-day SRT 4-day SRT 2-day SRT Figure 19. Percentage content of Volatile Suspended Solids Experiment 10 In the experimental reactor, both TSS and VSS concentrations were stable during the first 8-day-SRT period (days 0 through 24), with a slight decrease towards the end (day 22). Average solids concentration of 1,803 mg/L (TSS) and 1,655 mg/L (VSS) were observed during the 8-day-SRT period. These concentrations corresponded to 92% average content of VSS respect to the TSS concentration. Both TSS and VSS decreased significantly after the SRT was lowered to 6 days but were constant during the entire 6day-SRT period (days 25 through 42). This steady trend indicates that the solids concentrations were not seriously affected by the poor settling performance observed during this period. Average concentrations were 963 mg/L (TSS) and 918 mg/L (VSS), with an average VSS content of 95%, indica ting biomass growth as compared to the precedent period. This increment in growth can be explained by less competition for substrate and nutrients due to a lower biomass concentration. Therefore, solids concentration increased during the 4-day-SRT period (days 43 through 54) when, in addition, settling performance was recovered. Average concentrations during this period were 1,170 mg/L (TSS) and 1,075 mg/L (VSS) with an average VSS content of 92% similar to the first SRT tested (8-day-SRT). Finally, during the 2-day SRT period the average TSS (931 mg/L) and VSS (854 mg/L) concentrations decreased significantly due

PAGE 57

47 to the reduction of the SRT. However, the 92% (VSS/TSS) biomass content was maintained. These results indicate a good capacity of the system to overcome the pressure given by the reduction of the SRT. Table 10 compares the average concentra tions and percentage VSS/TSS between both reactors among the SRTs tested. Both reactors had similar TSS, VSS and VSS/TSS (%) values during the 8-day-SRT period (days 0 through 24). However, during the 6-day-SRT period, a significant difference was observed. The control reactor was significantly impacted by the poor settling performance and washout of biomass during this period (days 25 through 42), whereas the experimental reactor showed st able solids concentrations during this period. Despite the reduction of the SRT and the poor settling performance observed, the biomass concentration in the experimental reactor remained constant throughout the experiment, while the control reactor showed differences in each period. Table 10. Average Solids Concentrations and Biomass Content Experiment 10 Control Experiment SRT period TSS VSS VSS/TSS TSS VSS VSS/TSS 8 1,696 1,560 92% 1,803 1,655 92% 6 1,456 1,263 87% 963 918 95% 4 1,350 1,150 85% 1,170 1,075 92% 2 864 791 92% 931 854 92% 7.4.3 Specific Nitrate Formation Rates Exp. 10 Due to the difference in the solids concentration between reactors, specific nitrate formation rates were calculated by normalizing the nitrate formation rates to the VSS concentration. Hence, specific nitrate formation rates were expressed as mg NO 3 -N/g VSS-min. As shown in Figure 20, specific n itrate formation rates in the control reactor were consistently low throughout the experiment with a maximum rate of 0.003 mg NO 3 -N/g VSS-min (day 1) and an average rate of 0.001 mg NO 3 -N/g VSS-min. The minimum specific nitrate formation rate was observed by the end of the third 8-day-SRT (day 22) with a value of 0.0002 mg NO 3 -N/g VSS-min. During the complete 8-day-SRT period (day 0 through 24), the specific nitrate formation rates slightly declined until day

PAGE 58

24 when recovery was observed. The specific nitrate formation rates increased the first 6-day-SRT (days 25 through 30) until the second SRT was reached (day 32), when a reduction was observed again. These results were linked to a detriment in the settling performance occurred during the complete 6-day-SRT period (day 24 through 42). Steady, specific nitrate formation rates during the 4-day-SRT period (day 43 through 54) were observed, although the average rates for this particular period (0.0004 mg NO 3 -N/g VSS-min) were 33% of the average rate of the 8-day-SRT period (0.0012 mg NO 3 -N/g VSS-min), 50% the average rate of the 6-day-SRT period (0.0008 mg NO 3 -N/g VSSmin), and about 40% of the average specific nitrate formation rate for the experiment. Average, maximum, and minimum rates for each period (tripled SRTs) are listed in Table 11. Due to poor settling of the solids, nitrate formation rates in the control reactor during the 4-day-SRT period were significantly low (0.0003 mg NO 3 -N/g VSS-min). A slight increase was observed during the 2-day-SRT period (day 54 through 60), related to the improvement of settling performance. This period had an average rate of 0.0009 mg NO 3 -N/g VSS-min, similar to the average rate in the control reactor (0.001 mg NO 3 -N/g VSS-min). 0.000 0.004 0.008 0.012 0.016 0.020 024681012141618202224262830323436384042444648505254565860 Time (days)NO3 --N (mg/g VSS-min ) Control Experiment 8-day-SRT 6-day-SRT 4-day SRT 2-day SRT Figure 20. Specific Nitrate Formation Rates Experiment 10 48

PAGE 59

49 Table 11. Specific Nitrate Formation Rates during Different SRTs Exp.10 Control Experiment SRT (days) Period (days) Average Maximum Minimum Average Maximum Minimum 8 0 24 0.0011 0.0032 0.0002 0.0041 0.0092 0.0023 6 25 42 0.0008 0.0016 0.0003 0.0039 0.0060 0.0026 4 43 54 0.0004 0.0005 0.0003 0.0050 0.0053 0.0044 2 55 60 0.0009 0.0013 0.0005 0.0036 0.0044 0.0027 All 0 60 0.0009 0.0032 0.0002 0.0041 0.0092 0.0023 All rates are expressed in mg NO 3 -N/g VSS-min The experimental reactor showed a positive effect of CO 2 with a maximum specific nitrate formation rates of 0.006 mg NO 3 -N/g VSS-min (days 6 and 36), and an overall average of 0.004 mg NO 3 -N/g VSS-min. A minimum rate of 0.002 mg NO 3 -N/g VSS-min was observed on day 22 when settling performance began to deteriorate. The experimental reactor showed a steady performance with minor reductions on day 20 when poor settling performance was observed, and day 54 when the SRT was lowered to 2 days. A drastic decrease on specific nitr ate formation rates was observed during the first 4 days of the experiment due to the new conditions for the biomass (shorter SRT), although values stabilized from day 4 through th e end of the 8-day-SRT period (day 24). An average rate of 0.0036 mg NO 3 -N/g VSS-min was observed during this period (days 0 through 24), very similar to the overall average rate of the experimental reactor. During days 25 through 42 (6-day-SRT period), a steady performance was observed for the first 6-day SRT (day 25 through 30) followed by an increase in rates during the second 6-day SRT (day 31 to 36) when a steady performance was again maintained until day 50. An average rate of 0.0038 mg NO 3 -N/g VSS-min was observed during the 6day-SRT period, slightly higher than the precedent period and closer to the overall average rate (0.004 mg NO 3 -N/g VSS-min). A maximum rate of 0.0059 mg NO 3 -N/g VSS-min was reached by the second 6-day-SRT (day 36), corresponding also to the maximum rate reached throughout the experiment. This maximum rate was reached despite the poor settling performance during this period. The 4-day-SRT (day 43 through 54) period was mostly stable with the highest average rate (0.0048 mg NO 3 -N/g VSSmin) among the tested SRTs and with a maximum rate of 0.0052 mg NO 3 -N/g VSS-min

PAGE 60

50 (day 50), very similar to the maximum rate for the overall experiment (0.006 mg NO 3 N/g VSS-min). A significant recovery on settling performance favored the optimal specific nitrate formation rates during this particular period. However, a reduction of the nitrate formation rates was observed after day 50, when the SVI increased, until day 54 when settling performance was recovered again. Therefore, the 2-day-SRT period (days 54 through 60), showed moderately increasing rates associated with a better settling performance. This increment in rates can also be related to less competition for substrate (N) due to the lower solids (biomass) concentration. An average rate of 0.0036 mg NO 3 N/g VSS-min was observed during this period, which is similar to previous periods (8 and 6-day-SRT), and lower than the average rate obtained during its precedent period (4day-SRT). These results show that with the addition of 1% CO 2 during aeration, nitrification is possible at SRTs as low as 4 and 2 days with an efficacy similar to that obtained during 8, and 6-day-SRTs. By comparing the specific nitrate formation rates in both reactors, rates in the experimental reactor were found to be 6 times higher in average. A maximum ratio of 8.5 (experiment/control) was reached by the end of the second 4-day-SRT (day 50). Settling performance during this day was iden tical for both reactors, suggesting that differences in the specific nitrate formation rates were not only related to the settling performance but perhaps also to the conditions in the experimental reactor (1% CO 2 ) that might have selected for nitrifying bacteria. A minimum ratio of 2 (experiment/control) was observed during the first 8-day-SRT (days 4 and 8) and during the second 6-daySRT (day 32). Overall results validate hypothesis 1 of this research, which states that nitrification is possible at low SR Ts when high concentrations of CO 2 are supplied during aeration. 7.4.4 Settling Performance Exp. 10 The settling performance during Experiment 10 was evaluated with the SVI obtained as described in the materials a nd methods section (Equation 4). Figure 21, presents the SVI values obtained for both reactors. A red dashed line indicates the reference value below which settling performance is considered normal (SVI of 150 mL).

PAGE 61

Values of SVI above 150 mL indicate poor settling and the possible proliferation of filamentous bacteria. 0 100 200 300 400 500 600 700 024681012141618202224262830323436384042444648505254565860 Time (days)SVI (mL/g) Control Experiment Reference value 8-day SRT 6day 4-day SRT 2-day SRT Figure 21. Sludge Volume Index in Both Reactors Experiment 10. In the control reactor, a maximum SVI of 636 mL/g SS was observed on day 32, which affected the average SVI throughout the experiment (254 mL/g SS). After 6 days of operation, poor settling was observed in the control reactor throughout the experiment until day 50 when settling performance recovered. Washout of biomass was observed only on days 27 through 29 (less than 50 mL MLSS/cycle) and days 32 through 38 (~ 100 mL MLSS/cycle) with SVI values greater than 300 mL/g SS. These values are twice the values reported in the literature (Metcalf & Eddy 2003) for biomass washout, indicating that the SBR system used in this research had a better tolerance to poor settling than that of the actual full scale systems in WWTP. However, on day 30 (SVI: 504 mL/g SS) no washout of biomass was observed because the settled sludge in the reactor was just below the wasting line (located at 1 L mark from bottom to top) although fragmented settling was observed in the cylinder used for the SVI test as shown in Figure 22. During this period of poor settling, rising of the sludge blanket was observed due to poor N 2 bubbles produced during the anoxic periods. Additional problems possibly related to 51

PAGE 62

viscous bulking (suggested by the jelly-like appearance of the MLSS) were associated with high the SVI values observed and washout of biomass. The experimental reactor had an SVI average of 210 mL/g SS and a maximum of 446 mL/g SS (day 34). Settling performance was acceptable throughout the experiment except during the 6-day-SRT period (days 25 to 42) when poor settling and bulking problems were observed in both reactors. Although SVI values were above 150 mL/g SS during approximately 60% of the experiment washout of biomass and viscous bulking were not observed. Foaming observed duri ng the poor settling period disappeared when the SRT was lowered to 4 days (day 42). This can be explained by the slow growth of foam-causing microorganisms such as Nocardia and Microthrix (Pitt and Jenkins 1990), for which the reduction of the SRT might have worked as a selective pressure. In general, the overall settling performance of the experimental reactor was acceptable. EXP CON Figure 22. Sludge Blanket Test for Both Reactors Day 34, Experiment 10. Arrows indicate fragmented sludge blanket. 52

PAGE 63

53 By comparing both reactors, a better overall settling performance was observed in the experimental reactor as well as a better ability to recover from the reduction of the SRT. These results indicate a positive impact of the additional supply of CO 2 in the experimental reactor after the first 2 hours of the react cycle. 7.4.5 Ammonium Removal Exp. 10 Figure 23 shows the removal efficiency of NH 4 + and Figure 24 shows the final concentrations of NH 4 + -N in the supernatant. In the control reactor the removal efficiencies were greater than 90% throughout the experiment, except for day 3 (37%) when the NH 4 + concentration in the supernatant at the end of the cycle was 11.7 mg NH 4 + -N/L. This peak value was probably due to the adaptation of the ammonia oxidizing bacteria. After the completion of the first 8-day SRT (day 8) until day 20, the removal of NH 4 + was between 70 and 80% with a minimum of 55% (8.4 mg NH 4 + -N/L) observed on day 16, when the second 8-day SRT was completed. Concentrations of NH 4 + -N in the supernatant averaged 1.9 mg NH 4 + -N/L (for an average removal of 90%), and a minimum concentration of 0.5 mg NH 4 + -N/L was obtained 4 days after dropping the SRT to 6 days (day 28). Average concentrations in the supernatant for each particular period showed no significant difference among the 6, 4, and 2-day SRTs (0.9 mg NH 4 + -N/L in all cases) as shown in Figure 24. However, a higher average concentration in the supernatant (3.3 mg NH 4 + -N/L) during the 8-day-SRT period shows a s light effect of reducing the SRT on the removal of NH 4 +

PAGE 64

0% 20% 40% 60% 80% 100% 024681012141618202224262830323436384042444648505254565860 Time (days)Remova 54 l Control Experiment 6day 4-day SRT 2-day SRT 8-day SRT Figure 23. Removal Efficiency of Ammonium Experiment 10. 0.0 1.0 2.0 3.0 4.0 5.0 2 3 4 5 6 7 8 SRT (days)NH4 +-N (mg/L) Control Experiment Figure 24. Average Ammonium Concentra tion in Supernatant for different SRTs during Experiment 10

PAGE 65

55 In the experimental reactor, an average removal of NH 4 + of 93% and a maximum removal of 97% (0.5 mg NH 4 + -N/L) indicated good nitrogen removal of the system. Low removal of NH 4 + (70 80%) during the complete second 8-day SRT (day 8 through 16) was consistent with low removal in the control reactor due to the loss of settling performance. The lowest removal registered (71%) by the completion of the second 8day SRT (day 16) corresponded to a concentration in the supernatant of 5.4 mg NH 4 + N/L, which is above the MCL. The average supernatant concentrations for each particular SRT were very similar among the 6, 4, and 2-day-SRT periods (0.8, 0.9, and 0.9 mg NH 4 + -N/L respectively) differing only from the 8-day-SRT period for which the NH 4 + concentrations averaged 2.0 mg NH 4 + -N/L. A slight difference between both reactor s was observed only for the 8-day-SRT period with average NH 4 + concentrations in the supernatant of 3.2 and 2.0 mg NH 4 + -N/L in the control and experimental reactors respectively. However no significant difference was observed over time as shown in Figure 24, suggesting that higher concentrations of CO 2 do not have a significant positive or negative impact on ammonia oxidizing bacteria for any of the tested SRTs. 7.4.6 Nitrate Concentrations in Supernatant Exp. 10 Figure 25 illustrates the nitrate concentrations in the supernatant of both reactors. The control reactor showed an average concentration of 0.87 mg NO 3 -N/L, a maximum of 8.5 NO 3 -N mg/L (day 0), and a minimum of 0.3 mg NO 3 -N/L observed on days 3, 22, and 50 when nitrification rates were also low (less than 0.0003 mg NO 3 -N/L-min). Low concentrations of NO 3 -N in the supernatant were observed throughout the experiment in the control reactor consistent with the low nitrate formation rates obtained due to the short SRTs tested.

PAGE 66

0 1 2 3 4 5 6 7 8 9 10 024681012141618202224262830323436384042444648505254565860 Time (days)NO3 --N (mg/L) 56 Control Experim ent 6-day SRT 4-day SRT 2-day SRT 8-day SRT Figure 25. Average Nitrate Concentration in Supernatant Experiment 10. The experimental reactor had an average concentration of NO 3 -N in the supernatant of 1.58 mg NO 3 -N/L although this value was affected by the maximum concentration of 9.3 mg NO 3 -N/L (day 0) and a peak concentration of 6.7 mg NO 3 -N/L reached on day 3. A minimum concentration of 0.2 mg NO 3 -N/L was observed during day 2 due to the adaptation of the activated sludge to the new conditions. Nitrate concentrations in the supernatant were low throughout the experiment, with concentrations greater than 1.0 mg NO 3 -N/L only during the first 8 days of operation (first 8-day-SRT). Consistent with the n itrate formation rates obtained, there was no significant difference among the average values per SRT; as shown in Figure 26, indicating that similar performance can be reached at lower SRTs when 1% CO 2 is supplied after 2 hours of the beginning of the react cycle.

PAGE 67

0.0 1.0 2.0 3.0 4.0 5.0 2 4 6 8 SRT (days)NO3 --N (mg/L) Control Experiment Figure 26. Average Nitrate Concentration in Supernatant per SRT Exp.10. Low levels of nitrate in the supernatant for both reactors may have occurred due to nitrate utilization by filamentous bacteria during the settling period, which is also reflected on the high SVI values observed. Although nitrate concentrations for both reactors were low throughout the experiment, c oncentrations in the experimental reactor were twice in average the concentration in the control reactor, and a maximum ratio of 5 (experimental/control) was obtained. High concentrations of NO 3 -N in the supernatant are not desirable although nitrification rates are. Nitrate concentrations above 10 mg NO 3 -N/L were not observed, which is in compliance with the current MCL. 7.4.7 COD removal Removal efficiency of COD is shown in Figure 27 and remaining COD in the supernatant, measured as mg/L as O 2 is shown in Figure 28. As with previous experiments, an initial COD concentration of 267 mg/L as O 2 corresponding to two thirds of the COD in the synthetic wastewater (400 mg/L as O 2 ) was compared to the final concentrations in the supernatant to obtain the removal efficiencies. 57

PAGE 68

58 85% 90% 95% 100% 105% 024681012141618202224262830323436384042444648505254565860 Time (days)COD Removal Contro l Experi ment 8-day SRT 6-day SRT 4-day SRT 2-day SRT Figure 27. Removal Efficiency of COD Experiment 10 0 5 10 15 20 25 30 35 40 45 50 024681012141618202224262830323436384042444648505254565860 Time (days)COD (mg/L as O2) Control Experiment 8-day SRT 6-day SRT 4-day SRT 2-day SRT Figure 28. COD in the Supernatant Experiment 10 The removal efficiencies in the control reactor were always greater than 90%, with a maximum efficiency of 100% reached during days 3, 14, 25, and 26. An average removal efficiency of 95% was observed throughout the experiment and no significant

PAGE 69

59 differences were observed among the different SRTs although lower removals closer to 90% were observed towards the end. In all cases, final concentrations were below the MCL of 30 mg/L as O 2 with a maximum COD of 27 mg/L as O 2 observed on day 44 (2 days after the SRT was dropped to 4 days). The average remaining COD in the supernatant of the control reactor was 10 mg/L as O 2 The experimental reactor had an average COD removal of 96%, a maximum removal of 100%, and a minimum of 91%. A maximum COD of 24 mg/L as O 2 was reached on day 26 (2 days after the SRT was dropped to 6 days), and an average value of 10.8 mg/L as O 2 was observed throughout the experiment. According to the concentrations of COD in both reactors, no significant impact of CO 2 and low SRTs was observed in COD removal efficiencies and supernatant concentrations. Both reactors showed the same trends and had comparable values. Even though the experimental reactor showed slightly higher removal efficiencies, they were not significantly different. Both reactors met the minimum required removal efficiency of COD for secondary treatment (90%), and th e supernatant concentrations were always below the MCL (30 mg/L as O 2 ), indicating adequate performance of the system. 7.4.8 Summary of Results The results from Experiment 10 showed successful nitrification by supplying higher concentrations of CO 2 During this experiment, the performance of the activated sludge system (i.e. settling and COD removal) was not affected even though the system was operated at SRTs as low as 4 days. These are the first lab-scale results that suggest that activated sludge system can improve nitrate formation rates by providing CO 2 to a portion of the aeration basin. Overall results from Experiment 10 indicate a significant impact of CO 2 in specific nitrate formation rates at different SRTs, which validates hypotheses 1 and 2. Maximum nitrate formation rates for each reactor were 0.005 and 0.012 mg NO 3 -N/Lmin (control and experimental respectively), both slightly lower than those obtained during Experiment 9 (control: 0.004 mg NO 3 -N/L-min; experiment: 0.016 mg NO 3 -N/Lmin). Average nitrate formation rates were 0.001 and 0.005 mg NO 3 -N/L-min for the

PAGE 70

60 control and experimental reactors respectively. Both maximum and average nitrate formation rates in the experimental reactor were higher in all the experiments conducted, strengthening hypotheses 1 and 2. Moreover, after normalizing the rates of Experiment 10 to the VSS concentration, maximum, average, and minimum specific nitrate formation rates were still higher in the experimental reactor. Maximum specific nitrate formation rates were (0.003 and 0.009 mg NO 3 -N/L-min for control and experimental respectively) up to 3 times higher in the experimental reactor. However, on day 50, specific nitrate formation rates were up to 17 times higher in the experimental reactor. By comparing the average specific nitrate formation rates of each reactor (0.001 and 0.004 mg NO 3 -N/Lmin for control and experimental respectively), rates were 4 times higher in the experimental reactor. However, if the daily ratios (experiment/control) are averaged, an average ratio of 6 (experiment/control) is found. These results validate hypotheses 1 and 2, and are consistent with the findings of other researches, which have found a positive effect of CO 2 on nitrification rates and in the gr owth rate of nitrifiers (Gordon and Paskins 1982; Sakairi et al. 1996; Byong-Hee et al. 2000; Denecke and Liebig 2003; Wett and Rauch 2003). Although nitrate formation rates have not been reported, NO x -N generation (mg/L) was found to be approximately 3 times higher (1.5% CO 2 vs. 0% CO 2 ) after 2 hours of operation, similar to the results obtained during this research (Denecke and Liebig 2003). Additionally, Denecke and Liebig (2003) reported that the growth rate ( obs ) of mixed autotrophic and heterotrophic sludge increased 20% with a CO 2 concentration close to 1%. Other aut hors (Bringmann 1961; Gordon and Paskins 1982) also suggested a positive impact of CO 2 on the growth rates of nitrifying bacteria according to Denecke and Liebig (2003). However, availability of these papers was either limited or in a language other than English. Addition of 1% CO 2 during aeration after the 2 first hours of the react cycle led to better settling performance and faster recovery of the system after the reduction of the SRT. During Experiment 10, the experimental reactor did not presented washout of biomass although SVI values were greater than 150 mL/g 60% of the time, and a maximum SVI of 446 mL/g was reached on day 34. In contrast, the control reactor lost 5 10% of biomass/day during 10 days of the experiment, and showed a maximum SVI

PAGE 71

61 value of 636 mL/g on day 32. Additional layered settling was observed in the control reactor whereas the experimental reactor showed adequate settling characteristics. Comparison of the adequate settling performance observed in the experimental reactor during Experiments 9 and 10, with to the poor settling observed in the same reactor during Experiment 8, suggests a positive effect of CO 2 only when supplied after 2 hours of the react cycle. As discussed previously, these results are consistent with previous research that reported a type of heterotrophic filamentous bacteria (Thiothrix spp.) to be a able to synthesize inorganic carbon while in the presence of acetate (Odintsova et al. 1993; Nielsen et al. 2000). Furthermore, Wett et al (2003) found that the addition of bicarbonate (as CaCO 3 ) was an effective remedy against poor settling in activated sludge systems for the removal of nitrogen from wastewater with high concentrations of NH 4 + No significant positive or negative effect of CO 2 on the removal efficiency of COD was observed in any of the experiments conducted for this research. COD removal efficiencies and concentrations in the supernatant were within the limits for secondary treatment of wastewater. 7.5 Future Research The results obtained in this research led to several research opportunities including nitrification with low SRTs, optimum CO 2 concentrations to improve nitrification, effects of CO 2 on settling performance, and identification of CO 2 -sensitive bacteria. Some potential research topics are listed below based on these results and those reported in the literature. Although a positive effect of 1% CO 2 was observed on nitrification rates during this research, a range of pCO 2 can be evaluated by using the same experimental set-up to find the optimal CO 2 concentration at which maximum nitrification rates occur. Previous research suggest that concentrations of CO 2 higher than 1% can have inhibitory effects on nitrifying bacteria (Green et al. 2002; Denecke and Liebig 2003). However, these researchers did not include the evaluation of different SRTs, settling performance, and COD removal efficiencies.

PAGE 72

62 The effect of high concentrations of CO 2 supplied during aeration on ammonium oxidizing rates was inconclusive after this res earch. Therefore, further research including culture and identification of nitrifying bacteria using FISH is needed to determine if the growth rate of both ammonium and nitr ite oxidizing bacteria is affected by CO 2 Although several studies have been conducted to minimize the bulking problems in BNR systems, including reactor configurati on, aeration strategies (Noutsopoulos et al. 2002) and the use of selectors (Van Loosdrech t et al. 1998; Davoli et al. 2002), just few have tested the effect of inorganic carbon (Wett et al. 2003) on settling performance. However, these researches used CaCO 3 as source of inorganic carbon and suggested that dissociation of the Ca 2+ ions could have contributed to a better settling instead of the inorganic carbon. Therefore, and based on the results obtained in this research, further research on the effect of other source of inorganic carbon (i.e. CO 2 ) could be conducted to support previous research. The potential conflict that currently exists for the simultaneous removal of nitrogen and phosphorus, for phosphorus removal requires short SRTs as opposite to nitrogen removal (Van Loosdrecht et al. 1998), could be minimized based on the results of this research that indicate that nitrification in activated sludge systems can be achieved with short SRTs. Further research featuring the removal of both nutrients when supplying high concentrations of CO 2 could lead to a significant contribution to the BNR processes.

PAGE 73

63 8. Conclusions This research showed that successful nitr ification can be obtained at SRTs as low as 4 days without affecting the performance of the activated sludge system (i.e. settling and COD removal), when supplying higher concentrations of CO 2 during aeration. These are the first lab-scale results that suggest that activated sludge system can improve nitrate formation rates by providing CO 2 to the aeration basin. These findings can lead to change the current concept of nitrification as a slow process that requires more than 20 days to be completed. Moreover, they complement and support previous findings that suggested a positive effect of CO 2 on nitrification rates. Previous studies did not te st the effect of reducing the SRT. It was demonstrated that supply of 1% CO 2 during aeration in the activated sludge systems increases the average nitrate formation rates up to 4 times, makes nitrification possible at low SRTs, improves settling performance, and contributes to the optimal treatment performance of lab-scale activated sludge reactors operated as SBRs. The different effects of CO 2 on settling performance when supplied during the entire react cycle and when supplied after 2 hours of the beginning of the react cycle contribute to current research on bulking sl udge. On one hand, bulking occurred when CO 2 was supplied during the entire react cycle. This corroborates recent findings of bulking caused by obligate heterotrophic filamentous bacteria abundant in activated sludge that are suggested to be also a facultative chemolithoautotrophic type of filamentous bacteria (i.e. Thiothrix spp.). On the other hand, the improvement observed on settling performance when CO 2 was supplied after acetate was consumed (after 2

PAGE 74

64 hours from the beginning of the cycle) corroborat es the results of previous research that identified inorganic carbon as a potential remedy to poor settling and bulking sludge problems in activated sludge systems. In general, significant improvements to the nitrification process in activated sludge WWTPs can be derived from this research. Such improvements could include greater plant capacity, better understanding of the nitrification process, of the bulking problems, and of the microbial population that may be used to accelerate the treatment.

PAGE 75

65 9. References Byong-Hee, J., T. Yasunori and U. Hajime (2000). "Stimulating Accumulation of Nitrifying Bacteria in Porous Carrier by A ddition of Inorganic Carbon in a ContinuousFlow Fluidized Bed Wastewater Treatment Reactor." J Biosci Bioeng 89(4): 334-9. Dagley, S. and C. N. Hinshelwood (1938). "Physicochemical Aspects of bacterial Growth. Part II. Quantitative dependence of the Growth Rate of Bact. Lactis aerogenes on the Carbon Dioxide Content of the Gas Atmosphere." J Chem Soc 1938:1936-42 Davoli, D., P. Madoni, L. Guglielmi, M. Pergetti and S. Barilli (2002). "Testing the effect of selectors in the control of bul king and foaming in full scale activated-sludge plants." Water Sci Technol 46(1-2): 495-8. Denecke, M. and T. Liebig (2003). "E ffect of carbon dioxide on nitrification rates." Bioprocess Biosyst Eng 25(4): 249-53. Gaval, G., P. Duchene and J. J. Pernelle (2002). "Filamentous bacterial population dominance in activated sludges subject to stresses." Water Sci Technol 46(1-2): 49-53. Gordon, L. and A. Paskins (1982). "Influ ence of High Partial Pressure of Carbon Dioxide and/or Oxygen on Nitrification." J Chem Tech 32: 213 223. Grady, C. P. L. J., G. T. Daigger a nd H. C. Lim (1999). Biological Wastewater Treatment, Marcel Dekker.

PAGE 76

66 Green, M., Y. Ruskol, A. Shaviv and S. Tarre (2002). "The effect of CO2 concentration on a nitrifying chalk reactor." Water Res 36(8): 2147-51. Hall, S. J., J. Keller and L. L. Blackall (2003). "Microbial quantification in activated sludge: the hits and misses." Water Sci Technol 48(3): 121-6. Head, I. M., W. D. Hiorns, T. M. Embley, A. J. McCarthy and J. R. Saunders (1993). "The phylogeny of autotrophic ammoni a-oxidizing bacteria as determined by analysis of 16S ribosomal RNA gene seque nces." J Gen Microbiol 139 Pt 6: 1147-53. Jirka, A. M. and M. J. Carter (1975). "Micro semi-automated analysis of surface and wastewaters for chemical oxygen demand." Anal Chem 47(8): 1397-1402. Juretschko, S., G. Timmermann, M. Schmid, K. H. Schleifer, A. PommereningRoser, H. P. Koops and M. Wagner (1998). "Combined molecular and conventional analyses of nitrifying bacterium diversity in activated sludge: Nitrosococcus mobilis and Nitrospira-like bacteria as dominant populations." Appl Environ Microbiol 64(8): 304251. Kim, D. J., T. K. Kim, E. J. Choi, W. C. Park, T. H. Kim, D. H. Ahn, Z. Yuan, L. Blackall and J. Keller (2004). "Fluorescence in situ hybridization analysis of nitrifiers in piggery wastewater treatment reactors." Water Sci Technol 49(5-6): 333-40. Liu, J. R., C. A. McKenzie, E. M. Seviour, R. I. Webb, L. L. Blackall, C. P. Saint and R. J. Seviour (2001). "Phylogeny of the f ilamentous bacterium 'Nostocoida limicola' III from activated sludge." Int J Syst Evol Microbiol 51(Pt 1): 195-202. Martins, A. M., J. J. Heijnen and M. C. van Loosdrecht (2003). "Effect of dissolved oxygen concentration on sludge settl eability." Appl Microbiol Biotechnol 62(56): 586-93.

PAGE 77

67 Metcalf & Eddy, I. (2003). Wastewater E ngineering: Treatment and Reuse. New York, NY, McGraw-Hill. Mobarry, B. K., M. Wagner, V. Urbain, B. E. Rittmann and D. A. Stahl (1996). "Phylogenetic probes for analyzing abundance and spatial organization of nitrifying bacteria." Appl Environ Microbiol 62(6): 2156-62. Nielsen, P. H., M. A. de Muro and J. L. Nielsen (2000). "Studies on the in situ physiology of Thiothrix spp. present in ac tivated sludge." Environ Microbiol 2(4): 38998. Noda, N., N. Kaneko, M. Mikami, Y. Kimochi, S. Tsuneda, A. Hirata, M. Mizuochi and Y. Inamori (2003). "Effects of SRT and DO on N2O reductase activity in an anoxic-oxic activated sludge system." Water Sci Technol 48(11-12): 363-70. Noutsopoulos, C., D. Mamais and A. D. Andreadakis (2002). "The effect of reactor configuration and operational mode on Microthrix parvicella bulking and foaming in nutrient removal activated sludge systems." Water Sci Technol 46(1-2): 61-4. Odintsova, E. V., A. P. Wood and D. P. Kelly (1993). "Chemolithoautotrophic growth of Thiothrix ramosa." Arch Microbiol 160: 152-157. Peters, M., M. Newland, T. Seviour, T. Broom and T. Bridle (2004). "Demonstration of enhanced nutrient removal at two full-scale SBR plants." Water Sci Technol 50(10): 115-20. Pitt, P. and D. Jenkins (1990). "Causes and control of Nocardia in activated sludge." J Wat Polln Con Fedn 37(2): 151 162.

PAGE 78

68 Rittman, B. E. and P. L. McCart y (2001). Environmental Biotechnology: Principles and Applications. New York, NY, McGraw-Hill Companies, Inc. Sakairi, M. A. C., K. Yasuda and M. Matsumura (1996). "Nitrogen Removal in Seawater Using Nitrifying and Denitrifying Bacteria Immobilized in Porous Cellulose Carrier." Wat Sci Tech. 34(7-8): 267 274. Schramm, A., D. De Beer, M. Wagner a nd R. Amann (1998). "Identification and activities in situ of Nitrosospira and Nitrospira spp. as dominant populations in a nitrifying fluidized bed reactor." Appl Environ Microbiol 64(9): 3480-5. Tandoi, V., N. Caravaglio, D. D. D. Balsamo, M. Majone and M. C. Tomei (1994). "Isolation and physiological characterization of Thiothrix sp." Wat Sci Technol 29: 261 269. U.S. EPA (1993). "Nitrogen Cont rol Manual." (EPA/625/R-93/010). U.S. EPA (1999). Wastewater Technology: Sequencing Batch Reactors. W. D. C. Office of Water, U.S. Environmental Protection Agency EPA. Van Loosdrecht, M. C. M., and Jetten, M.S.M. (1998). "Microbiological Conversions in Nitrogen Removal." Wat Sci Technol 38(1): 1-7. Van Loosdrecht, M. C. M., F. A. Brands e and A. C. de Vries (1998). "Upgrading of Wastewater Treatment Processes for Integrated Nutrient Removal The BCFS Process." Wat Sci Tech. 37(9): 209-217. Wagner, M., G. Rath, H. P. Koops, J. Flood and R. Ammann (1996). "In Situ Analysis of Nitrifying Bacteria in Sewage Treatment Plants." Wat Sci Technol 34(1-2): 237-244.

PAGE 79

69 Wett, B., A. Eladawy and W. Becker (2003). "Carbonate addition--an effective remedy against poor activated sludge settling properties and alkalinity conditions in small wastewater treatment plants." Water Sci Technol 48(11-12): 411-7. Wett, B. and W. Rauch (2003). "The role of inorganic carbon limitation in biological nitrogen removal of extremely ammonia concentrated wastewater." Water Res 37(5): 1100-10.

PAGE 80

70 10. Bibliography Metcalf & Eddy, Inc (1991). Wastewater Engineering: Treatment, Disposal, and Reuse. 3 rd edition. Published by McGraw Hill, U.S.A., 1991. pp 431 Pirt, S.J. (1975). Principles of microbe and cell cultivation. Published by John Wiley & Sons, Inc., New York, 1975. pp.77,78

PAGE 81

71 Appendices

PAGE 82

72 Appendix A: Calibration of the Ion Selective Electrodes Calibration of the ammonium and nitrate ion selective electrodes was performed daily, 1 hour before the beginning of the first react cycle. The calibration procedures specified by the manufactures were modified due to a significant difference between the ionic strength of the MLSS (approx. 0.02 M) and the ionic strength recommended by the manufacturer to prepare the standard solutions (0.12 M for the ammonium electrode and 0.1 M for the nitrate electrode). Table A-1 presents the ions in solution according to the composition of the synthetic wastewater (Table 3), along with the ion concentrations (M) and the product of the concentration of each ion and the square of their correspondent charge. Table A-1. Ions Present in the Synthetic Wastewater Ion Charge Concentration (M) Concentration*(charge^2) (BO 3 ) -3 -3 2.43E-07 2.18E-06 (C 2 H 3 O 2 ) -1 1.04E-02 1.04E-02 Ca +2 2 1.26E-04 5.05E-04 Cl -1 2.75E-03 2.75E-03 Co +2 2 6.30E-08 2.52E-07 Cu +2 2 1.92E-06 7.69E-06 Fe +2 2 5.40E-06 2.16E-05 H + 1 7.28E-07 7.28E-07 HPO 4 -2 -2 4.84E-04 1.94E-03 I -1 1.81E-08 1.81E-08 K + 1 4.83E-04 4.83E-04 Mg +2 2 3.65E-04 1.46E-03 Mn +2 2 7.58E-06 3.03E-05 Na + 1 1.23E-02 1.23E-02 NH 4 + 1 2.00E-03 2.00E-03 SO 4 -2 -2 1.01E-03 4.03E-03 Zn +2 2 2.30E-06 9.18E-06

PAGE 83

73 Appendix A: (Continued) The ionic strength (I) of the synthetic wastewater was computed as half the sum of the product of the ions concentrations and the squared charge. An ionic strength of 0.02 M was obtained. A.1. Calibration of the Ammonium Probe Two ammonium combination glass body electrodes (Cole-Parmer 27502-03, Cole-Parmer Instrument Company, Vernon Hills, IL) and ion meters (Oakton Benchtop Ion 510 Meter and Oakton Ion 6 Meters, Oakton Instruments, Vernon Hills, IL) were calibrated daily, 1 hour before the beginning of the first react cycle. The ammonium electrodes used a 0.1M NaCl filling solution (Cole Parmer 27503-78 reference filling solution, Cole-Parmer Instrument Company, Vernon Hills, IL). A1.1 Solutions for the Calibration of the Nitrate Probes A 1000 ppm NH 4 + -N Ammonium standard solution was prepared in the laboratory by mixing 2.97 grams of reagent-grade NH 4 Cl in 1 L of DI water. The Ionic Strength Adjuster (ISA) used was a solution of 5M NaCl prepared in the laboratory by mixing 292 grams of reagent-grade NaCl in 1 L of DI water. Table A-2 presents a range of standard solutions with concentrations of 1, 10, 20, 30, and 40 ppm NH 4 + -N, obtained from diluting the stock ammonium standard solution (1000 ppm NH 4 + -N). The I of standard was adjusted to 0.02 M by adding 0.2 mL of ISA. Table A-2. Standard Solutions for the Calibration of the Ammonium Probes Concentration D.I Water Volume of Standard Solution (1000 ppm NH 4 + N) ISA 0.1 ppm NH 4 + -N 50 mL 0.005 mL 0.2 mL 1 ppm NH 4 + -N 50 mL 0.05 mL 0.2 mL 10 ppm NH 4 + -N 50 mL 0.5 mL 0.2 mL 20 ppm NH 4 + -N 50 mL 1.0 mL 0.2 mL 30 ppm NH 4 + -N 50 mL 1.5 mL 0.2 mL 40 ppm NH 4 + -N 50 mL 2.0 mL 0.2 mL

PAGE 84

74 Appendix A: (Continued) Similarly, Table A-3 presents several volumes of the ammonium standard solution added to 50 mL samples of regular feed, to obtain feed samples with concentrations of 1, 10, 20, and 30 ppm of NH 4 + -N. No ISA was added to these samples since the feed already had an ionic strength of approximately 0.02 M. These feed samples with different concentrations of ammonium were used to compare the probe readings when using the standard solutions with the readings when using the actual feed. Table A-3. Feed Samples with Variable Nitrate Concentrations for the Calibration of the Ammonium Probes Concentration Feed 1 Volume of Standard Solution (1000 ppm NH 4 + -N) 1 ppm NH 4 + -N 50 mL 0.05 mL 10 ppm NH 4 + -N 50 mL 0.5 mL 20 ppm NH 4 + -N 50 mL 1.0 mL 30 ppm NH 4 + -N 50 mL 1.5 mL 1 Feed was added to the standard solution to complete a volume of 50 mL A dilution of the standard solution (1000 ppm NH 4 + -N) with a concentration of 0.01 M NH 4 Cl was used for overnight storage of the nitrate probes. Fresh storage solution was made daily. A1.2 Calibration Protocol for the Ammonium Probes The Calibration of the ammonium probes was performed by using the solutions shown in Tables A-2 and A-3. The protocol was as follows: 1. Probes were immersed in DI water for 10 minutes 2. Probes were immersed in 0.1 ppm NH 4 + -N Standard Solution for 10 minutes

PAGE 85

75 Appendix A: (Continued) 3. Probes were sequentially immersed in dilutions of 1, 10, 20, 30, and 40 ppm NH 4 + -N (Table A-2). The probes were kept immersed for 2 minutes in each dilution and readings in mV were recorded from the meters after 2 minutes before switching the probes to a solution with a higher nitrate concentration. The ammonium probes were rinsed with DI water before being immersed into a new dilution. During the 2-minute period, the solutions were kept mixed with a magnetic stirrer at a velocity of 3 rpm. 4. After the 20 ppm NH 4 + -N dilution, the probes were immersed for 10 minutes in DI water. 5. Probes were immersed in 0.1 ppm NH 4 + -N Standard Solution for 10 minutes 6. Probes were immersed in feed samples of 1, 10, 20, and 30 ppm NH 4 + -N (Table A-3) in the order mentioned for 2 minutes, rinsing the probes with DI water between dilutions. The readings in mV were taken after 2 minutes. Figure A1-1 presents a calibration curve prepared for each ammonium probe during a typical day of Experiment 10 (Day 2). The readings in mV collected during calibration had an exponential fit. The equations of the trendlines were used to determine the ammonium concentration in NH 4 + -N from the mV readings.

PAGE 86

Appendix A: (Continued) y = 44.02e0.0502xR2 = 0.9999 y = 45.166e0.0473xR2 = 0.9999 1 10 100 -100 -80 -60 -40 -20 0 Meter Readings (mV)NH4 +-N (mg/L) Ammonium Probe 1 Ammonium Probe 2 Figure A-1. Ammonium Probes Calibration Curves Tables A-4 presents the readings in mV and the correspondent concentrations obtained during calibration of the nitrate probes with the standard solutions on day 7. Table A1.5 presents the readings in mV and the computed nitrate concentrations for the different feed samples checked. The % error between the concentrations obtained with the standard solutions and the concentrations obtained with the feed samples was equal or lower than 10%. If the error was greater than 10% recalibration was required. Table A-4. Calibration of Ammonium Pr obes with Standard Solutions (Day 2) Ammonium Probe 1 Ammonium Probe 2 Dilution (mg/L NH 4 + -N) mV mg/L NH 4 + -N mV mg/L NH 4 + -N 1 -75.2 1.0 -80.3 1.0 10 -29.8 9.9 -32.5 9.7 20 -15.9 19.8 -17.1 20.1 30 -7.6 30.1 -8.5 30.2 40 -1.6 40.6 -2.4 40.3 76

PAGE 87

77 Appendix A: (Continued) Table A-5. Ammonium Concentration in Feed Samples during Calibration (Day 2) Ammonium Probe 1 Ammonium Probe 2 Dilution (mg/L NH 4 + -N) mV mg/L NH 4 + N Error mV mg/L NH 4 + N Error 1 -75.4 1.0 0% -79.0 1.1 7.6% 10 -30.7 9.4 5.7% -33.2 9.4 -6.1% 20 -15.7 20.0 0.1% -17.6 19.6 -1.8% 30 -7.1 30.8 2.7% -8.7 29.9 -0.2% A.2. Calibration of the Nitrate Probe Two nitrate combination glass body electrodes (Cole-Parmer 27502-03 and Nitrate combination glass body electrode, Cole-Parmer 27502-31, Cole-Parmer Instrument Company, Vernon Hills, IL) and ion meters (Oakton Ion 6 Meter, Oakton Instruments, Vernon Hills, IL) were calibrated daily. The nitrate electrodes used a 0.1M (NH 4 ) 2 SO 4 filling solution (Cole Parmer 27503-79 reference filling solution, ColeParmer Instrument Company, Vernon Hills, IL) A.2.1 Solutions for the Calibration of the Nitrate Probes A 1000 ppm NO 3 -N nitrate standard solution was prepared in the laboratory by mixing 1.37 grams of reagent-grade NaNO 3 with 1 L of DI water. The ISA had a concentration of 1M NaSO 4 and was prepared in the laboratory by mixing 119 grams of NaSO 4 with1 L of DI water. Table A-6 presents a range of standard solutions with concentrations of 1, 5, 10, 15, and 20 ppm NO 3 -N, obtained from diluting the stock nitrate standard solution (1000 ppm NO 3 -N). The Ionic Strength of each diluted standard was adjusted to 0.02 M by adding 0.667 mL of ISA (per 50mL of sample) as indicated in Table A1-6.

PAGE 88

78 Appendix A: (Continued) Table A-6. Standard Solutions for the Calibration of the Nitrate Probes Concentration D.I Water 1 Volume of Standard Solution (1000 ppm NO 3 -N) ISA 0.1 ppm NO 3 -N 50 mL 0.005 mL 0.667 mL 1 ppm NO 3 -N 50 mL 0.05 mL 0.667 mL 5 ppm NO 3 -N 50 mL 0.25 mL 0.667 mL 10 ppm NO 3 -N 50 mL 0.5 mL 0.667 mL 15 ppm NO 3 -N 50 mL 0.75 mL 0.667 mL 20 ppm NO 3 -N 50 mL 1.00 mL 0.667 mL 1 DI water was added to the standard solution to complete a volume of 50 mL Similarly, Table A-7 presents several volumes of the nitrate standard solution added to 50 mL samples of regular feed to obt ain feed samples with concentrations of 1, 5, 10, 15, and 20 ppm of NO 3 -N. No ISA was added to the samples since the feed already had an ionic strength of approximately 0.02 M. These feed samples with different concentrations of nitrate were used to compare the probe readings when using the standard solutions with the readings when using the actual feed. Table A-7. Feed Samples with Variable Nitrate Concentrations for the Calibration of the Nitrate Probes Concentration Feed 1 Volume of Standard Solution (1000 ppm NO 3 -N) 1 ppm NO 3 -N 50 mL 0.05 mL 5 ppm NO 3 -N 50 mL 0.25 mL 10 ppm NO 3 -N 50 mL 0.5 mL 15 ppm NO 3 -N 50 mL 0.75 mL 1 Feed was added to the standard solution to complete a volume of 50 mL A dilution of the standard solution (1000 ppm NaNO 3 ) with a concentration of 0.01 M NaNO 3 was used for overnight storage of the nitrate probes. Fresh storage solution was made daily.

PAGE 89

79 Appendix A: (Continued) A.2.2 Calibration Protocol for the Nitrate Probes The Calibration of the nitrate probes was performed by using the solutions shown in Tables A1-6 and A-7. The protocol was as follows: 1. Probes were immersed in DI water for 10 minutes 2. Probes were immersed in 0.1 ppm NO 3 -N Standard Solution for 10 minutes 3. Probes were sequentially immersed in dilutions of 1, 5, 10, 15 and 20 ppm NO 3 -N (Table A-6). The probes were kept immersed for 2 minutes in each dilution and readings in mV were recorded from the meters after 2 minutes before switching the probes to a solution with a higher nitrate concentration. The nitrate probes were rinsed with DI water before being immersed into a new dilution. During the 2-minute period, the solutions were kept mixed with a magnetic stirrer at a velocity of 3 rpm. 4. After the 20 ppm NO 3 -N dilution, the probes were immersed for 10 minutes in DI water. 5. Probes were immersed in 0.1 ppm NO 3 -N Standard Solution for 10 minutes 6. Probes were immersed in feed samples of 1, 5, 10 and 15 ppm NO 3 -N (Table A-7) in the order mentioned for 2 minutes, rinsing the probes with DI water between dilutions. The readings in mV were taken after 2 minutes. Figure A-2 presents a calibration curve prepared for each nitrate probe during a typical day of Experiment 10 (Day 7). The equations of the trendlines were used to determine the nitrate concentration in NO 3 -N from the mV readings.

PAGE 90

Appendix A: (Continued) y = 786436e-0.0572xR2 = 0.9997 y = 111283e-0.0436xR2 = 0.9997 1 10 100 180 200 220 240 260 280 Meter Reading (mV)NO3 --N (mg/L) Nitrate Probe 1 Nitrate Probe 2 Figure A-2. Nitrate Probes Calibration Curves Tables A-8 presents the readings in mV and the correspondent concentrations obtained during calibration of the nitrate probes with the standard solutions on day 7. Table A1.5 presents the readings in mV and the computed nitrate concentrations for the different feed samples checked. The % error between the concentrations obtained with the standard solutions and the concentrations obtained with the feed samples was equal or lower than 10%. Errors greater than 10% required recalibration. Table A-8. Calibration of Nitrate Probes with Standard Solutions (Day 7) Nitrate Probe 1 Nitrate Probe 2 Dilution (mg/L NO 3 -N) mV mg/L NO 3 -N mV mg/L NO 3 -N 1 266 1.0 237 1.0 5 230 4.9 209 5.1 10 214 9.9 197.5 9.8 15 204 15.3 189.8 15.2 20 197.3 20.4 184.4 20.6 80

PAGE 91

81 Appendix A: (Continued) Table A-9. Nitrate Concentration in Feed Samples during Calibration (Day 7) Nitrate Probe 1 Nitrate Probe 2 Dilution (mg/L NO 3 -N) mV mg/L NO 3 -N %Error mV mg/L NO 3 N %Error 1 265 1.1 7% 236 1.1 8% 5 231 4.7 -6% 210 4.8 -5% 10 214 9.9 -1% 198 9.6 -4% 15 204 15.3 2% 189.9 15.1 0%