ADVANTAGES AND DISADVANTAGES OF MICROPOROUS MEMBRANES IN A HOLLOW FIBER BIOREACTOR FOR SPACE APPLICATIONS by MARIA NOEL RUIZ CARERI, B.S.Ch.E. A THESIS IN CIVIL ENGINEERING Submitted to the Graduate Faculty of Texas Tech University in Partial Fulfillment of the Requirements for the Degree of MASTER OF SCIENCE IN CIVIL ENGINEERING Approved Audra Morse Chairperson of the Committee Andrew Jackson Accepted John Borrelli Dean of the Graduate School August, 2005
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
ADVANTAGES AND DISADVANTAGES OF MICROPOROUS
MEMBRANES IN A HOLLOW FIBER BIOREACTOR
FOR SPACE APPLICATIONS
by
MARIA NOEL RUIZ CARERI, B.S.Ch.E.
A THESIS
IN
CIVIL ENGINEERING
Submitted to the Graduate Faculty of Texas Tech University in
Partial Fulfillment of the Requirements for
the Degree of
MASTER OF SCIENCE
IN
CIVIL ENGINEERING
Approved
Audra Morse Chairperson of the Committee
Andrew Jackson
Accepted
John Borrelli Dean of the Graduate School
August, 2005
ii
ACKNOWLEDGMENTS
I would first like to thank my major advisor, Dr. Audra Morse, who guided me
through my graduate studies at Texas Tech University. I appreciate her constant support,
patience, and understanding during my thesis writing and all the help she provided me to
complete my degree. Also, my appreciation goes to Dr. Jackson, for participating on my
committee, for his review of this thesis, and also for giving me the opportunity to obtain a
remarkable work graduate experience.
Thanks to the Texas Tech Athletic Department, without them I would never have
had the opportunity to be in this country. I am very grateful to my co-workers and
friends who have been with me in the last couple of years, and without the TTU-NASA
research team this research would not have been possible. Thanks to Dr. Dallas and the
people in electrical engineering for taking their time to help me out, to Dr. Heyward
Ramsey for his encouragement and trust in the past couple of years, and to Eric
McLamore, for his patience and understanding on those crazy days where nothing made
sense, for always giving me a good laugh.
And of course, special thanks go to my family, especially my parents, who have
always been there to support me and taught me to never quit. Also, to my sister who
gave me the confidence to finish my degree and succeed in anything I do. They have
played the most important role in forming the person I am today.
iii
TABLE OF CONTENTS
ACKNOWLEDGMENTS ii
ABSTRACT vii
LIST OF TABLES viii
LIST OF FIGURES xi
LIST OF ABBREVIATIONS xiii
CHAPTER
I. INTRODUCTION 1
II. BACKGROUND 5
2.1Water in Space 5
2.2 Hollow Fiber Membrane Bioreactors 6
2.2.1 Membrane Types and Geometry 7
2.2.2 Membrane Modes for Aeration 9
2.3 Nitrification 10
2.4 Transport Processes 14
2.4.1 Hydrodynamics 16
2.4.2 Mass Transfer 16
2.4.3 Dimensionless Groups 19
2.5 Biofilms 20
2.5.1 Biofilm Development 21
2.5.2 Biofilm in MABRs 21
iv
III. MATERIALS AND METHODS 23
3.1 Membrane Module 23
3.2 HFMBR 25
3.2.1 Inoculation 25
3.2.2 Feed Composition 26
3.3 HFMBR Reactor System 27
3.3.1 System Components 27
3.3.1.1 Pressure Measurement 27
3.3.1.2 Connections 27
3.3.2 Analytical Methods 29
3.4 Loading Studies 29
3.5 Nitrification 30
3.6 Hydrodynamics/Tracer Studies 31
3.7 Mass Transfer Experiments 32
3.8 Biofilm Analysis 33
3.8.1 Determination of biofilm distribution throughout the bioreactor 35
3.8.1.1 Circumferential Analysis 35
3.8.1.2 Longitudinal Analysis 35
3.8.1.3 Outside-in Strategy 36
3.8.2 Identification of the bacteria present within the biofilm 37
v
3.8.2.1 Identification of heterotrophic bacteria by cultivation in nutrient
agar 38
3.8.2.2 Identification of Nitrobacter species by AT5N Medium 38
3.8.2.3 Identification of Nitrosomonas Europaea by AT5N Medium 39
IV. RESULTS 41
4.1 Nitrification 41
4.2 Loading Studies 48
4.3 Hydrodynamics 55
4.4 Mass Transfer 61
4.5 Biofilm Analysis 70
4.5.1 Visual Analysis 71
4.5.2 Circumferential Analysis 75
4.5.3 Longitudinal Analysis 76
4.5.4 Biofilm Thickness 77
4.5.5 Bacteria Identification 82
V. CONCLUSIONS 84
REFERENCES 88
APPENDICES
A. RAW DATA DURING THE HFMBR STUDY 94
B. RAW DATA. SIGMA PLOT GRAPHS 114
C. MASS TRANSFER CALCULATIONS 118
vi
D. TRACER STUDY CALCULATIONS 128
E. T-TEST 135
vii
ABSTRACT
Texas Tech University (TTU) works in conjunction with NASA to develop a
wastewater recovery system robust enough for use on long term-space missions.
Biological treatment has been the primary focus at TTU, with specific thrusts in
developing a biological treatment system that may be operated with minimal crew
maintenance and low energy and mass requirements.
Hollow fiber membrane bioreactors (HFMBRs) may be used for biological
wastewater treatment, and may be integrated with NASA’s current research
developments. The goal of this research is to (a) evaluate the effect of mass transfer by
the use of microporous membranes and their application for microgravity conditions; (b)
compare the effect of membrane type and configuration on treatment efficiency to
previous literature values; and (c) determine the amount and distribution of biofilm
growth within the reactor. Therefore, the objective of this research was accomplished
using a microporous HFMBR.
From the experimental studies performed for this thesis it was found that the
HFMBR exhibits promising use in space applications. Maximum nitrification efficiency
at low loading rates and high HRTs were accomplished using the HFMBR. Therefore,
characteristics such as, suitable bioreactor size and the efficiency obtained during its
operation, the HFMBR offer a potential for NASA’s needs; nonetheless, developing a
system with more favorable system hydrodynamics would aid to improve treatment
efficiency in a HFMBR
viii
LIST OF TABLES
2.1 Dimensionless numbers 20
3.1 HFMBR Characteristics 25
3.2 Composition of HFMBR feed 26
3.3 Nutrient agar composition 38
3.4 AT5N medium composition (Nitrobacter) 39
3.5 AT5N Medium Composition (Nitrosomonas) 40
4.1 Evolution of nitrification process 43
4.2 Effluent results for HFMBR during +4NH -N loading 44
4.3 Ammonium-nitrogen removal rate from various treatment systems 47
4.4 Mass removed at a constant rate 54
4.5 Dispersion in the HFMBR 59
4.6 HFMBR mass transfer without biofilm 62
4.7 Oxygen transfer within the HFMBR 63
4.8 HFMBR mass transfer after biofilm formation 65
4.9 Dimmensionless numbers 67
4.10 Comparison of mass transfer correlations from literature 70
4.11 Cross-sectional mass per surface area 75
4.12 Longitudinal mass per surface area 77
4.13 Biofilm thickness 78
A1 Test point 1. Influent Concentrations (mg/L) 95
ix
A2 Test point 1. Effluent Concentrations (mg/L) 96
A3 Test point 2. Influent Concentrations (mg/L) 97
A4 Test point 2. Effluent Concentrations (mg/L) 98
A5 Test point 3. Influent Concentrations (mg/L) 99
A6 Test point 3. Effluent Concentrations (mg/L) 100
A7 Test point 4. Influent Concentrations (mg/L) 101
A8 Test point 4. Effluent Concentrations (mg/L) 102
A9 Test point 5. Influent Concentrations (mg/L) 103
A10 Test point 5. Effluent Concentrations (mg/L) 104
A11 Test point 6. Influent Concentrations (mg/L) 105
A12 Test point 6. Effluent Concentrations (mg/L) 106
A13 Test point 7. Influent Concentrations (mg/L) 107
A14 Test point 7. Effluent Concentrations (mg/L) 108
A15 Average Values 109
A16 Standard Deviation Values 110
A17 Volumetric Conversion Rates 111
A18 Mass Calculations 112
A19 Reactor Efficiency using variable surface areas 112
A20 Sample Calculations 113
C1 Initial DO calculations at 0.3 mL/min 119
C2 Initial DO calculations at 5 mL/min 120
C3 Initial DO calculations at 10 mL/min 121
x
C4 Initial DO calculations at 21 mL/min 122
C5 Initial DO calculations. Summary table 123
C6 Final DO calculations at 0.3 mL/min 124
C7 Final DO calculations at 1 mL/min 125
C8 Final DO calculations at 10 mL/min 126
C9 Final DO calculations. Summary table 127
C10 Biofilm consumption 127
D1 Initial tracer at 0.3 mL/min 129
D2 Initial tracer at 1 mL/min 130
D3 Initial tracer at 15 mL/min 131
D4 Final tracer at 0.3 mL/min 132
D5 Final tracer at 1 mL/min 133
D6 Final tracer at 15 mL/min 134
xi
LIST OF FIGURES
2.1 Mass Transfer in a nitrifying microporous HFMBR 17 3.1 Hollow fiber membrane reactor (HFMBR) schematic 24
3.2 Hollow fiber membrane reactor system layout 28
3.3 HFMBR module 28
3.4 Sampling locations within the HFMBR 34
3.5 Different cross-sections within the HFMBR 36
3.6 Membrane sampling 37
4.1 HFMBR conversion rates 42
4.2 HFMBR efficiency 45
4.3 Volumetric conversion rates 49
4.4 Conversion rates per surface area 50
4.5 Overall reactor performance 51
4.6 Conversion rate comparison 53
4.7 Comparison of the HFMBR removal efficiencies 55
4.8 Hydrodynamic experiments without biofilm 57
4.9 Hydrodynamic experiments after biofilm growth 58
4.10 Oxygen transfer within the HFMBR 64
4.11 Sherwood vs. Reynolds number. Before and after biofilm formation 69
4.12 Sampling locations within the HFMBR 71
4.13 Biofilm growth within different locations of the HFMBR 73
xii
4.14 Outside-in strategy 74
4.15 Biofilm distribution within the HFMBR 76
4.16 Biofilm thickness 78
4.17 Biofilm thickness at sections 2A and 3A 80
4.18 Biofilm thickness at sections 4A and 5A 81
B1 Raw data for pH and TDS 115 B2 Raw data for dissolved oxygen and ammonia 116 B3 Raw data for nitrogen oxides 117
xiii
LIST OF ABBREVIATIONS
d Unitless dispersion number
D Diffusion coefficient
DDI Deionized, distilled water
DO Dissolved oxygen
HFMBR Hollow fiber membrane bioreactor
HRT Hydraulic retention time
J Flux
JE Experimental flux
JM Flux consumed by microorganisms
KL Mass transfer coefficient in the liquid phase
KO Overall mass transfer coefficient
MABR Membrane aerated bioreactor
+4NH -N Ammonium
−2NO -N Nitrite
−3NO -N Nitrate
−xNO -N Nitrogen oxide
Pe Peclet number
Re Reynolds number
Sc Schmidt number
Sh Sherwood number
1
CHAPTER I
INTRODUCTION
From the early 1970s to about 1980, the primary worries for wastewater treatment
were aesthetic and environmental concerns as indicated by the reduction of biological
oxygen demand, total suspended solids, and pathogenic organisms (Metcalf & Eddy,
2003). Later, the removal of nutrients began to be addressed. As time passed, the goals,
and objectives became, and still are, more strictly related to the water quality.
Wastewater treatment methods involve physical, chemical, or biological
reactions. However, wastewater treatment is not only applied in terrestrial environments,
but also wastewater treatment is a concern in space. Long duration space missions will
require the reuse of water supplies.
NASA is currently evaluating different physical and chemical methods for the
recovery of wastewater. A Water Recovery System (WRS) is necessary for long-term
space missions due to the limited capacity for water storage due to weight and volume
requirements. It is definitely a challenge for the WRS to produce stable and healthy
water from a wastewater stream composed of gray-water (hand, body, clothes, and dish
washing), humidity condensate, and urine. Biological wastewater treatment methods, as
well as the physical and chemical methods, also provide the possibility of organic carbon
removal, as well as ammonia and nitrate from wastewater. Biological wastewater
methods have the advantage of low energy requirements, no additional chemical
requirements for the treatment process, and the production of less waste material
2
requiring storage and handling. Therefore, biological treatment is a cost-effective
method for the replacement of the physical and chemical techniques currently considered
in space.
NASA’s Johnson Space Center (JSC) and Texas Tech University (TTU) have
been working simultaneously with the purpose of developing a robust system to be
operated under microgravity conditions and cost effective enough to be used for long-
term missions. Texas Tech University (TTU) currently operates several biological
systems for wastewater treatment (Morse et al., 2003; Jackson et al., 2004; Muirhead, et
al., 2003; and McLamore, 2004) including: the TTU-WRS, two membrane aerated
bioreactors, referred to as MABRs, a commercial hollow fiber polypropylene
microporous membrane bioreactor (HFMBR), and a dual system that promotes
simultaneous nitrification and denitrification (sAMR). In the TTU-WRS system,
nitrification is promoted in a tubular reactor located downstream of a packed bed reactor,
in which denitrification occurs. Nitrification refers to the conversion of ammonium to
nitrite and nitrate under aerobic conditions, while denitrification is the conversion of
nitrite and nitrate into nitrogen gas under anaerobic conditions (Jackson et al., 2004; and
McLamore et al., 2004). This system presented several maintenance and performance
problems; therefore, the design and treatment efficiency of a membrane-aerated
bioreactor to treat NASA's simulated wastewater was considered.
MABRs are hollow fiber membrane reactors that consist of permeable tubes
through which oxygen is diffused into the system in order to support biofilm growth.
MABRs present several advantages over the tubular reactor such as bubble-free aeration,
3
low solids production, high resistance to shock loadings, and low maintenance (Morse et
al., 2003). However, effective membrane type and configuration has not been
established. For example, membranes may be placed in parallel or randomly distributed
within the reactor. Parallel membranes refer to a straight configuration where the
membranes do not come in contact with each other, while on the other hand, random
membranes refer to the arbitrarily distribution of the membranes in the reactor.
Research at TTU focuses on the use of silicone membranes randomly distributed
within MABRs. The selection of random membranes over parallel membranes was due
to the assumption that by using random membranes mixing would be enhanced;
therefore, increasing transport and improving overall reactor performance. However, the
most effective membrane type and configuration has not been established. Results show
that random membranes could increase transport but certain difficulties may be faced due
to the application of this type of packing configuration, presenting a drawback to their
application. Bao and Lipscomb (2002) analyzed the effect of packing configuration on
mass transfer, and results indicated reduction in mass transfer due to channeling
(formation of biofilm) in randomly distributed fibers. Also, the use of microporous
membranes could contribute to a more efficient treatment; however, their use could
present a problem when applied under microgravity conditions. Silicone membranes can
be operated at higher pressures without forming bubbles, while the formation of bubbles
is possible in microporous membranes due to the porosity of the membranes (Ahmed and
Semmens, 1992).
4
Hollow fiber membrane bioreactors are most commonly used for filtration
purposes. Membranes have been used to remove contaminants from wastewater in
microfiltration and ultrafiltration. For systems previously used, hollow fiber membranes
have also been used for gas stripping. By applying a vacuum through the membranes,
volatile organic compounds have been removed from wastewater streams. However, the
use of HFMBRs for stripping does not present the same problems as when used for
aeration purposes. Some of the anticipated problems presented when using HFMBRs for
aeration purposes are short circuiting as a result of biofilm growth in between the
membranes and the lack of flow control since the use of a vacuum is not applied.
Although some difficulties will arise, it is believed that the use of microporous
membranes, in comparison to silicone membranes, would increase transport of oxygen
through the membranes, improving the treatment efficiency within HFMBRs. Therefore,
the HFMBR is a possible candidate for the replacement of the tubular reactor. However,
all previous studies performed on HFMBRs have been with silicone and randomly
distributed membranes.
Thus there is a need to evaluate the effect of membrane type and configuration in
overall treatment efficiency of the reactor. The objectives of this thesis are to analyze the
advantages and disadvantages of microporous membranes in a hollow fiber membrane
bioreactor by (a) evaluating the effect of mass transfer by the use of microporous
membranes and their application for microgravity conditions; (b) comparing the effect of
membrane type and configuration on treatment efficiency to previous literature values;
and (c) determining the amount and distribution of biofilm growth within the reactor.
5
CHAPTER II
BACKGROUND
2.1 Water in Space
By the advance of science, humans have been able to explore space. Today, long-
term missions are not self sufficient. In long-term space missions, such as a trip to Mars,
astronauts need to perform the same activities in a space shuttle, without gravity, and
reduced space, but they still need to eat and drink. Therefore, the National Aeronautics
and Space Administration (NASA) has been working to develop a wastewater treatment
system for potential use in space applications.
Water is one of the most crucial provisions astronauts need to live and work in
space. That is why NASA has been working in developing physical, chemical or
biological methods to be applied to recycle the wastewater with the ultimate goal of
reducing the cost of missions by decreasing the payload weight. NASA’s focus in the
past had been physical and chemical systems to recycle water. However, NASA is
concentrating in developing biological processes for space applications.
NASA’s first four manned spaceflight projects were Mercury, Gemini, Apollo,
and SkyLab. In the past, water was generated by fuel cells that were used to provide
energy for the spacecraft and potable water was generated as a by-product, and water has
been recycled by physicochemical processes. A Water Recovery System (WRS) needs to
be 100% efficient, self-sufficient and capable of operation in microgravity conditions.
The use of a biological process for space applications may be energy efficient and
6
self-sufficient, require little or no maintenance in order to minimize the crew’s time for
other tasks, require little to no chemicals, and have low mass.
There are two main goals to be accomplished when operating a biological system;
the removal of organic constituents as well as the removal of nitrogen compounds. The
earliest version of a biological WRS was an immobilized cell bioreactor, conducted at
Johnson Space Center (JSC) in 1997 (Pickering et al., 1997). Also, other systems have
been previously analyzed (Finger et al., 1999; Petersen et al., 1991) that included
membrane technologies to develop membrane bioreactors. It is important to remember
that biological processes are followed by physiochemical to complete the water recycling
process, but they are not discussed in this thesis. The scope of this thesis is to evaluate a
membrane bioreactor for its application in space.
2.2 Hollow Fiber Membrane Bioreactors
The use of membranes for treatment of water and wastewater has increased in the
last several years. Membrane aerated bioreactors are hollow fiber membrane bioreactors
(HFMBRs), used most commonly for filtration, that represent a new technology for
aerobic wastewater treatment. Advantages such as bubble free aeration, low solids
production, high resistance to shock loadings, high nitrification efficiency, low
maintenance and a decrease in space requirements can be achieved by the use of
HFMBRs. In a hollow fiber bioreactor, oxygen flows through the lumen side of the
hollow fibers and oxygen diffuses through the wall of the membrane. Oxygen is utilized
by the bacterial population attached to the surface of the membranes, creating a driving
7
force for mass transfer. The membranes provide high oxygen permeability, ensuring the
transport of oxygen through the membranes and providing surface area for biofilm
attachment and treatment to occur.
The most important benefit obtained from HFMBRs is higher mass transfer.
Casey et al. (1999), establishes that the main advantage in process performance between
HFMBRs and conventional reactors is the active layer of biofilm formed on the
membranes and the importance of the active layer location. Depending on the waste
stream, most HFMBRs, consist of aerobic nitrifying bacteria located on the outside of the
film close to the membrane walls where oxygen is being provided and the anaerobic
denitrifying bacteria is located on the inside of the film where there is high organic
matter; therefore, dual mass transfer of oxygen and nutrients occurs from the inside and
outside of the membranes.
2.2.1 Membrane Types and Geometry
Membrane types and geometry are some important features to consider in order to
develop a low mass and energy efficient biological system. Selection of an appropriate
membrane is perhaps the most important feature. Membranes made of teflon, silicone,
gore-tex, polyetherimide, and silicone with fibrous support, silicone and polypropylene
have been used in the past for the removal of different pollutants such as synthetic
sewage, food processing wastewater, organic carbon and inorganic nitrogen between
others (Torrey, 1984). The configurations of membranes used include tubular, plate and
frame, single tube, tubular coil, and hollow fiber (Casey et al., 1999).
8
Aeration within a reactor is achieved by the use of membranes. Pressure provides
a gradient to encourage aeration through the membranes. The pressure driven
membranes are divided into three divisions based on membrane surface. These
membranes can be microporous, dense (silicone), and composite (dense coats on
microporous membranes). Membrane characteristics such as mass transfer, permeability,
pressure limitations, and membrane life span are some factors to take in consideration.
Microporous membranes present the advantage of having negligible resistance to
mass transfer; the bubble free form of aeration results in near 100 percent mass transfer
(Grimberg et al., 2000). Mass transfer takes place by diffusion through the pores of the
membrane. Nonetheless, pressure regulations are restricted due to the formation of
bubbles. The disadvantage of microporous membranes is the limitation to operating
pressure for which a pressure difference across the membrane of 2 to 3 psi was found to
cause bubbles (Ahmed and Semmens, 1991). If liquid penetrates into the micropores of
the membranes, reduction of mass transfer is observed and bubbles can be produced. The
formation of bubbles may present a drawback for the application of this type of
membranes under zero-gravity conditions. The life span of this type of membrane is
reduced due to the deposition of suspended solids and oils within the pores (Casey et al.,
1999) and they cannot be found in small diameter, and are relatively expensive.
Recent studies have investigated the replacement of porous membranes with
silicone membranes for wastewater applications (Ahmed and Semmens, 1992; Brindle
and Stephenson, 1996). Transport in dense membranes occurs via diffusion due to a
pressure differential. Oxygen has a high solubility in silicone; therefore, most dense
9
membranes are made of silicone. Dense membranes present several advantages over
microporous membranes, such as the use of high intramembrane oxygen pressures (up to
3*105 Pa) (Casey et al., 1999), high resistance to chemical and mechanical stress due to
the absence of pores, and the reduction of membrane fouling. In comparison to
microporous membranes (only when bubbles are formed), silicone membranes can also
operate at higher pressures without bubble formation, generating higher mass transfer
rates. Overall, dense membranes have better oxygen mass transfer with membrane
aeration than bubble aeration and have higher life span than microporous membranes.
2.2.2 Membrane Modes for Aeration
Membranes can be operated in two different modes. The dead-end mode, where
the membrane is pressurized with gas and one end of the fibers is sealed. In flow through
mode, gas is continuously pumped through hollow fibers and is vented to keep the partial
pressure of oxygen high along the membrane. The advantages of the dead end mode are
that the release of gases to the atmosphere is avoided and that 100 % gas transfer is
obtained since the only way the gas escapes is through diffusion through the membranes.
On the other hand, the disadvantages of this operation mode are condensation of water
inside the fiber membranes and the use of low pressures in order to obtain bubble less
aeration affecting the mass transfer rate (Ahmed and Semmens, 1992).
When operating membranes in flow through mode, vapor condensation is avoided
inside the membrane fibers and higher pressures can be utilized increasing the mass
transfer rate of a system. Since the gas is vented, complete transfer efficiency may not be
10
achieved and volatile organic compounds (VOC’s) may be stripped and vented to the
atmosphere. The VOC emissions present a concern due to their harmful effect to the
environment and the environmental compliance management costs. The process of
removing the VOCs from the environment is usually more expensive and troublesome
than avoiding the initial release of VOCs. Also, VOCs emissions present a definitely
unsafe environment in space.
HFMBRs have replaced conventional reactors for wastewater applications. The
presence of biofilm offers a higher rate of removal by the HFMBRs in comparison to
conventional treatment. Higher oxygen conversion when used with sealed end
membranes and high organic carbon removal rates can be achieved. The oxygen
diffusion rate is about 10 g/m2-d in conventional reactors while in HFMBRs up to 20
g/m2-d can be achieved (Torrey, 1984). HFMBRs can be used for simultaneous
nitrification and organic removal in a single reactor. Hollow fiber bioreactors are suitable
for simultaneous carbon substrate oxidation, nitrification (oxygen rich side of biofilm),
and denitrification (oxygen depleted biofilm) (Timberlake et al., 1988). However, a
microporous HFMBR for the sole purpose of nitrification was under scrutiny to complete
the objectives of this thesis.
2.3 Nitrification
Nitrification is a microbial process by which reduced nitrogen compounds
(primarily ammonia) are sequentially oxidized to nitrite ( −2NO ) and nitrate ( −
3NO ). This
is predominantly an aerobic chemoautotrophic process (Maier et al., 2000). Nitrifiers are
11
obligate aerobes that utilize oxygen (O2) for respiration and utilize inorganic carbon as an
energy source.
Nitrification is a two step process, typically involving two different types of
nitrifiers in the conversion of ammonia to nitrite and nitrate. True nitrifying bacteria are
considered to be those belonging to the family nitrobacteraceae. These bacteria are
strictly aerobic, gram-negative, chemolithic autotrophs. They require oxygen, utilize
mostly inorganic (without carbon) compounds as their energy source, and require carbon
dioxide (CO2) for their source of carbon. The energy sources are derived from the
chemical conversion of ammonia to nitrite, or nitrite to nitrate. Five genera are generally
accepted as ammonia-oxidizers and four genera as nitrite-oxidizers. Of these,
Nitrosomonas (ammonia-oxidizers) and Nitrobacter (nitrite-oxidizers) are the most
frequently identified genus (Watson et al., 1981).
In nitrification, first the oxidation of ammonia to −2NO (Eq. 2.1) is performed by
the Nitrosomonas, followed by the oxidation of −2NO to −
3NO (Eq. 2.2) by the
Nitrobacter species (Rittmann et al., 1994). For complete nitrification to occur, two
reactions must take place. Equation 1 shows the oxidation of ammonium to nitrite, and it
can be observed that two acid equivalents (H+) are created per mole of nitrogen oxidized.
Equation 2.2 shows the complete nitrification or conversion of the intermediate product
−2NO to −
3NO . Oxygen is required for the oxidation of ammonium and is used as the
terminal electron acceptor by the nitrifying bacteria.
12
+4NH + 1.5O2 → −
2NO + H2O + 2H+ Equation 2.1
∆G0 = -45.79 kJ per e- eq
−2NO + 0.5O2 → −
3NO Equation 2.2
∆G0 = -37.07 kJ per e- eq
The compound that gets oxidized is called the reductant and the substance that
gets reduced is called the oxidant. The oxidant is O2 and the reductant is +4NH ,
respectively (Rittmann and McCarty, 2001). Therefore, +4NH is the electron donor,
losing electrons to the electron acceptor O2, which gains an electron during the
nitrification process. When there is enough oxygen present, nitrification goes to
completion yielding −3NO ; however, an intermediate product is obtained when oxygen is
limiting in the reaction. It is important to remember that oxygen is not the only limiting
condition for nitrification, but that the carbon substrate can be limiting as well (Casey et
al., 1999). However, it can be observed from Equations 1 and 2, that the oxidation of
ammonium to nitrite is more energetically favorable in comparison to the second step in
nitrification. As a result, nitrite is fully consumed by bacteria in their environment
leading to predominantly the existence of nitrate (Rittmann et al., 1994).
Nitrification is an aerobic process that can occur not only in natural environments,
such as in lakes and rivers, but nitrification can be used for wastewater applications.
Nitrification can be accomplished in suspended or attached growth systems (Metcalf &
Eddy, 2003). Systems such as trickling filters, activated sludge, rotating biological
contactors, and packed beds have been used in the past. However, this research focuses
13
on nitrification occurring in attached growth systems, using membranes to increase
surface area per unit volume as composed to traditional attached growth systems.
The efficiency of the nitrification process is affected by the environmental
conditions (whether in environmental habitats or in a reactor), such as temperature, pH,
alkalinity, dissolved oxygen (DO), and nutrient availability (Udert et al., 2003). It has
been established that the rate of nitrification increases with increasing temperature.
Nitrification rates have been found to double for every 10°C increase in temperature
between 10°C and 30°C. According to Environmental Protection Agency (EPA)
findings, the pH levels below < 5.0, as well as pH > 8.0 have been reported to decrease
the rate of ammonium oxidation, decreasing the nitrification rates. Performance stability
is maintained at pH levels between 6.5 and 8.0. For complete nitrification to occur, the
amount of oxygen required is more than 4.57 g O2/g N, and certain wastewater
characteristics are necessary. At low DO concentrations (0.5 to 2.5 mg/L), nitrification
becomes limited (Metcalf & Eddy, 2003). A wastewater with low levels of organic
matter and the need of other micronutrients are also necessary in small amounts (P, S,
and Fe) for complete nitrification to occur.
In addition, factors such as membrane type and organic and hydraulic loading
have an effect on nitrification as well. Microporous membranes present higher surface
area per unit volume than silicone membranes. Synthetic membranes are thin, solid-
phase barriers that allow the passage of certain substances under the influence of a
driving force. Both the chemical and the physical nature of the membrane material
control membrane separation. Membrane separation occurs because of differences in
14
size, shape, chemical properties, or electrical charge of the substances to be separated.
Microporous membranes control separation by size, shape and charge discrimination,
whereas nonporous membranes depend on sorption and diffusion (Singh, 1998).
Nitrification depends on the surface area available for the attached
microorganisms, which are responsible for the conversion of ammonium to nitrate.
Nitrification is also dependent on the membrane permeability, which is dependant on the
membrane type for the diffusion of oxygen; in this case the electron acceptor. The
hydraulic loading, the rate at which the microorganisms are fed is also an important
factor to consider for high nitrification efficiency. These factors will be discussed in this
paper in subsequent sections.
Biological nitrogen removal by hollow fiber membrane bioreactors (HFMBRs) is
a promising method to remove nitrogen from wastewater. Nitrification in HFMBRs
occurs when the carbon substrate loading rate of the wastewater is low, and high oxygen
concentrations at the membrane-biofilm interface would support nitrification. A HFMBR
is used to accomplish nitrification for the removal of wastewater contaminants and the
possible use of this membrane process under microgravity conditions.
2.4 Transport Processes
There are three fundamental principles of transport processes. These mechanisms
are momentum transfer, heat transfer, and mass transfer. Momentum transfer is
concerned with the transfer of momentum, which occurs in moving media. Heat transfer
is concerned with the transfer of heat from one point to another, while mass transfer
15
involves the transfer of mass from one phase to another distinct phase (Geankoplis,
1983).
To determine a reactor’s performance, information on thermodynamics, physical
properties, hydrodynamics, and mass transfer must be known. In a HFMBR,
simultaneous mass transfer occurs. Gas diffuses through the membranes due to a
pressure differential, while at the same time, diffusion within the biofilm occurs due to
convective flow of the bulk liquid in the shell side of the reactor. Therefore, the overall
process involves diffusion of the gas through the membrane, transport from the bulk
liquid to the biofilm surface, diffusion through the biofilm, and transport of the liquid
from one point to another. Thus, the processes governing mass transport in HFMBRs are
mass transfer and hydrodynamics. The thermodynamic properties are assumed to be in
equilibrium.
In HFMBRs, the presence of packing provides a resistance to the flow of the fluid
that is greater than it would be in an empty column shell (Strigle, 1987). The non-
uniform distribution of the packing has an effect on liquid distribution of the flow as well
as gas velocity. It has been investigated in the past that the liquid flow is an active
element affecting the internal transport process in the anaerobic part of the biofilm
(Alphenaar et al., 1993); however, this has not been established due to several factors
affecting the hydrodynamics within the reactor. Parameters such as the mixing, residence
time distribution, and the influence of hydrodynamics on mass transfer are some factors
to take in consideration. Therefore, hydrodynamics and mass transfer are dependant on
each other.
16
2.4.1 Hydrodynamics
Mass transfer is influenced by the thickness of the membrane wall, the actual pore
diameter in the lumen, and the hydraulic flow through the membranes. The hydraulic
flow characteristics of complete-mix and plug-flow reactors can be described as varying
from ideal and non ideal, depending on the relationship of the incoming flow to outgoing
flow (Metcalf & Eddy, 2003).
A tracer may be used to recognize the hydraulic performance of a reactor. The
effect of short circuiting, channeling, flow patterns, and the actual residence time due to
biofilm growth can be determined by analyzing the tracer response curves. Moreover,
the tracer response curves may be used to estimate the biomass growth rates and the mass
transfer within the biofilm.
2.4.2 Mass Transfer
For a membrane system, the mass transfer process is determined by three mass
transfer resistances in series, resistance in the gas phase, the resistance due the membrane
and the resistance in the liquid. A HFMBR with microporous membranes was operated
for the purpose of this thesis. The membrane and gas resistances are considered much
smaller than the liquid resistance, thus neglected in the analysis for this report (Cote,
1989). Only the liquid mass transfer resistance is taken in consideration.
In HFMBRs simultaneous mass transfer occurs when oxygen diffuses through the
membranes within the reactor due to a pressure differential, and by forced convection
when nutrients are transported from the bulk liquid in the shell side of the reactor. Figure
17
2.1 shows the concentration profiles in the gas-membrane-liquid interfaces. Mass
transfer resistances are dependent on the hydrodynamic properties of the liquid phase and
packing structure. Resistances are smaller at larger, turbulent flows than at laminar
flows. Since mass transfer is dependent on the hydrodynamic conditions, and the
hydrodynamic conditions affect the biofilm growth within the HFMBR, the biofilm
thickness and structure affect the rate of mass transfer; therefore, affecting the overall
performance of the system.
Figure 2.1 Mass transfer in a nitrifying microporous HFMBR
Diffusion can be explained by Fick’s law and a mass transfer coefficient. Fick
proposed a linear relation between the rate of diffusion of a chemical species and the
local concentration gradient of that species (Cussler, 2002). The flux needs to be
18
calculated in order to determine the overall mass transfer coefficient. The flux, from
Equation 2.3 can be defined as the amount transferred per unit time, the flux (J).
)( cockJ −= Equation 2.3
where; J = flux of chemical at interface [M/V-T]
k = mass transfer coefficient [L/T] c = concentration of specie at interface [M/V] co = bulk concentration [M/V]
The flux, J, in Equation 2.3, includes both diffusion and convection. Mass
transfer across an interface is described in terms of a flux. Fick’s first law defines the
overall mass transfer coefficient as the sum of the three resistances in series, where the
value of each resistance is represented by its respective mass transfer coefficient. A
concentration differential is assumed and defined in Equation 2.4.
LMGO kkkk1111
++= Equation 2.4
where; 1/ Ok = overall mass transfer resistance [L/T]-1
1/ Gk = mass transfer resistance in the gas phase [L/T] -1 1/ Mk = mass transfer resistance through the membrane [L/T] -1 1/ Lk = mass transfer resistance in the bulk liquid [L/T] -1
The mass transfer resistance in the gas phase is determined by diffusion, and the
mass transfer in the liquid phase is determined by convection. Mass transfer has a strong
dependence on the biofilm structure, the hydrodynamic conditions, and substrate loading
on the biofilm surface (Viera et al., 1993). As previously stated, in a microporous hollow
fiber membrane reactor, the membrane and gas resistances are much smaller than the
19
liquid resistance, and considered negligible in mass transfer analyses of this type (Cote,
1989).
Additionally, Fick’s law takes in consideration the length and number of
membranes where diffusion is taken place; thus, the length and number of membranes
will be taken in consideration. From Equation 2.4 a new term, a diffusion coefficient (D)
and the membrane length (lm) are introduced.
lmcocDJ )( −
= Equation 2.4
where;
J = flux of chemical [M/V-T] D = diffusion coefficient [L2/T]
Table 3.1 HFMBR characteristics Membrane Characteristics Porosity 40% Porosity OD/ID 300µm OD/220µm ID Potting Material Epoxy Number of Fibers 3600 Maximum Temperature/Pressure 2.8 kg/cm2 (2.8 bar, 40 psig) at 23 °C with appropriate
hose clamps. Maximum 30 °C at lower pressures Active Surface Area 0.5 m2 (5.4 ft2) Priming Volume (ID) 63 mL Housing Characteristics Material Polysulfone Flange Connections Shellside (Gas/Vacuum) Standard Female Luer Lock Supplied with two ⅛ inch
Hosebarb adaptors which mate to 1/4 inch ID tubing Lumenside (Wetted Surface) 1/2 inch Hosebarb Weight Dry 0.15 kg (0.32 lbs.) Liquid Full (Lumenside) 0.2 kg (0.44 lbs.) Shiping Weight 0.3 kg (0.66 lbs.)
3.2 HFMBR
3.2.1 Inoculation
Inoculation of the reactor was completed using nitrifying bacteria batch culture
originally obtained from the Texas Tech University-Water Recovery System. The
nitrifying bacteria were grown and acclimated in the batch culture. A known volume of
deionized, distilled water (DDI) was added as well as nutrients such as ammonium
chloride (NH4Cl), sodium bicarbonate (NaHCO3) and Winogradskys Medium Modified
(Atlas, 1995). The Winogradskys Medium Modified was selected for cultivation of
26
nitrifiers. The changes in the ammonia and nitrogen oxides ( −xNO -N, −
2NO -N, −3NO -N)
concentrations were monitored. Ammonia oxidation activity was observed by a decrease
in the ammonia concentration and an increase in the −xNO -N concentration, indicating the
presence and growth of nitrifiers. The batch was fed once or twice a week, depending on
the rate of ammonium conversion.
After assuring the presence of nitrifiers in the batch culture, the mixed solution
was added to the HFMBR. The reactor was continually fed with NH4Cl, NaHCO3, and
the Winogradskys Medium Modified. The inorganic feed was to support an autotrophic
population within the reactor. Continuous monitoring of the ammonium and nitrogen
oxides persisted, as well as other conditions suitable for the growth of nitrifiers, such as
pH and dissolved oxygen within the bioreactor.
3.2.2 Feed Composition
An inorganic solution was used to feed the HFMBR. The feed was prepared daily
and consisted of sodium bicarbonate, ammonium chloride, DDI, and a solution similar to
batch culture feed, in proportions presented in Table 3.2. The feed tank was kept
homogeneous using a stirring bar.
Table 3.2 Composition of HFMBR feed Ingredient Amount [g/L]
DDI 1 NaHCO3 1.25 NH4Cl 0.625
Winogradsky's Solution 20
27
3.3 HFM Reactor System
3.3.1 System Components
The HFMBR consisted of a pump, influent and effluent tanks, silicone tubing
lines to connect the system, a mass flow controller and a pressure gage to control the air
pressure into the system. Figure 3.2 gives a simplified layout of the operated HFMBR.
3.3.1.1 Pressure Measurement
Oxygen was delivered by the addition of air to one of the cavities in the reactor
(Figure 3.3). Air was supplied from TTU facilities. A mass flow controller
manufactured by Cole Parmer (Model A-32464-16) was used to maintain a constant air
pressure within the system, while a digital pressure gage (Cole Parmer model HW-
68920-00) was used to just measure the pressure. Air was supplied to the reactor at a gas
pressure of 3.44 kPa (0.5 psi). Air flow was opposite to the water flow.
3.3.1.2 Connections
Feed and effluent tanks were connected to the reactor by the use of silicone
tubing (peroxide) obtained from Cole Parmer Instrument Co. The feed was delivered to
the reactor by 4 feet (MASTERFLEX®, L/S ™ 14) silicone tubing. The effluent line,
with the same characteristics presented above was of a length of 1 foot. The lines were
periodically cleaned and replaced to ensure that biofilm growth did not occur in the lines
but within the reactor.
28
Figure 3.2 Hollow fiber membrane reactor system layout
Figure 3.3 HFMBR module
29
3.3.2 Analytical Methods
Daily influent and effluent water samples were taken from the reactor and
N, total dissolved solids (TDS), and dissolved oxygen. All nitrogen measurements are
reported as mg-N/L.
Dissolved oxygen concentrations were measured by using a ROSS probe
(ThermoOrion 9708). All samples were filtered through sterilized membranes with
0.45 µm pore size and samples were prepared following Standard Methods (APHA,
1998). Filtrates were tested for nitrate ( −3NO -N), nitrite ( −
2NO -N) and ammonia (NH3-
N). The samples were analyzed by ion chromatography (DX-600) Dionex, USA) for the
detection of −3NO -N and −
2NO -N, a TOC machine manufactured by Shimadzu (Model
TOC-V CSH) for the detection of TOC, while NH3 was analyzed by using a ROSS probe
(ThermoOrion 9708).
3.4 Loading Studies
The performance of the HFMBR was related to the +4NH -N loading rates for
nitrification. Different loading rates were used to determine the reactor’s maximum
nitrification efficiency. The maximum nitrification efficiency within the HFMBR was
determined by analyzing the concentrations of NH3-N and −xNO -N (in mg/L) in the
influent and effluent of the system.
30
The nitrifying HFMBR was operated for approximately eight months. The feed
concentration was kept constant throughout the HFMBR operation; however, the loading
(flow) rate, at which the feed (mass ammonia fed to nitrifiers) was applied to the reactor,
changed over time. The HFMBR was operated at seven different hydraulic retention
times (HRT) including 0.15, 0.09, 0.07, 0.06, 0.04, 0.03, and 0.026 days, respectively.
After reaching steady state conditions, the loading rate (mg/day) was changed.
The need to keep the loading rates within the oxygen transfer capabilities of the
system was necessary for successful treatment of the HFMBR. The best removal
efficiency is expected at high HRTs (low flow rates) where there is enough contact time
for nitrification to occur. On the other hand, poor efficiency is expected at low HRTs
(high flow rates) where the microorganisms have little contact time for the removal of
NH3-N from the system.
Factors such as the hydrodynamics within the HFMBR influence treatment as
well as the biofilm growth and thickness in the reactor; for simplicity all these factors are
assumed to be constant throughout the experimental procedure. The reactor optimum
nitrification efficiency was therefore determined under two conditions: (1) by
determining the best loading rate for nitrification to occur, and (2) by determining the
effect of biofilm growth on mass transfer limitations.
3.5 Nitrification
Nitrification rates can be calculated by the ammonia removal rates, and the
amount of nitrite and nitrate produced. The reactor was fed with inorganic carbon to
31
support a pure culture of autotrophic bacteria; therefore, the heterotrophic population was
considered negligible. A nitrogen mass balance of the system was performed to calculate
the nitrification efficiency of the HFMBR. The ammonia oxidation percentage was
obtained by dividing the effluent nitrite plus nitrate concentration by the influent
ammonia concentration. The system performance, for −2NO -N, −
3NO -N, and ammonium
( +4NH -N) concentrations, was compared at different hydraulic retention times.
Nitrification within the system was calculated for each test point. Daily
measurements for influent NH3-N and effluent −xNO -N ( −
2NO -N + −3NO -N)
concentrations were considered. The measured NH3-N concentrations were converted to
+4NH -N in order to estimate the nitrification efficiency. Percent nitrification was
calculated using Equation 3.1.
(inf) N-NH
)eff( N-NO
4
-x
+=ionNitrificat Equation 3.1
3.6 Hydrodynamics/Tracer Studies
Tracer studies are useful to evaluate the hydraulic performance of a reactor.
Experiments were conducted on the reactor during the initial (no biofilm present) and
final (biofilm present) phases. Sodium bromide was used as a conservative tracer in a
continuous input tracer study. Sodium bromide was injected into the system at different
flow rates. Collection of influent and effluent samples continued until the effluent
bromide concentration matched the influent bromide concentration. Samples were
32
analyzed using automated ion chromatography (IC) and all values were reported as
(mg-Br/L).
The use of a tracer was used to recognize the dynamic behavior of fluid flow
through the HFMBR. The effect of short circuiting, channeling, flow patterns, and the
actual residence time due to biofilm growth were determined by analyzing the tracer
response curves. Moreover, the tracer response curves may be used to estimate the
biomass growth rates and the mass transfer within the biofilm. The number of
membranes in the reactor has an effect on the active surface area for biofilm growth
influencing the treatment efficiency. Replicates of the experiments were performed by
using different fluid flows. Hydrodynamic analyses were evaluated by using the data
collected during tracer studies at the initial (no biofilm present) and final (biofilm
present) stages of the bioreactor operation.
3.7 Mass Transfer Experiments
Oxygen was required as the terminal electron acceptor for respiration of the
microorganisms. The number of membranes in the reactor has an effect on the active
surface area for biofilm growth, directly influencing treatment efficiency. A negligible
membrane resistance for oxygen transfer is assumed due to the micro-porous size of the
membranes. Therefore, the only resistance to mass transfer is provided by the bulk
liquid. Previous research, done by Yang and Cussler (1989), had already established that
the majority of the resistance when using microporous membranes, was from the bulk
liquid.
33
To calculate the mass transfer within the hollow fiber membrane bioreactor
(HFMBR), aeration experiments were conducted. DDI was boiled and sparged with
nitrogen gas until the dissolved oxygen (DO) concentration was below 1 mg/L. The
influent and effluent DO concentrations were measured using a ROSS probe
(ThermoOrion 9708). The gas pressure was kept constant at 0.5 psi. Experiments were
conducted until saturation was reached. Experiments were conducted at different flow
rates and they were performed prior to biofilm formation, as well as after biofilm growth.
The effect of biofilm growth on mass transfer was then determined.
From the oxygen transfer experiments, the oxygen flux through the membranes,
mass transfer coefficients, and the diffusion coefficient were calculated and compared to
other researchers.
3.8 Biofilm Analysis
Biofilm growth was observed in the bioreactor. Determination of biofilm
distribution, thickness, and the determination of the bacteria present within the bioreactor
were objectives of this research. By analyzing the biofilm distribution throughout the
bioreactor, the hydrodynamics may be better understood.
To determine biofilm growth within the reactor, two different tasks were
accomplished: (1) the determination of biofilm distribution throughout the bioreactor, and
(2) the identification of the bacteria present within the biofilm. Sample strategies to
extract the biofilm attached to the membranes were developed.
34
The purpose of determining the biofilm distribution throughout the bioreactor was
to identify the preferential flow path, and predict the effect of biofilm formation on
hydrodynamics. A visual analysis was first completed. Biofilm distribution was
analyzed by obtaining the mass per surface area in the longitudinal direction and at four
different locations within the HFMBR circumference (Figure 3.4). Biofilm sampling was
accomplished in three different ways, in the longitudinal and circumferential direction
and from the outside towards the inside of the reactor, respectively.
Figure 3.4 Sampling locations within the HFMBR
35
3.8.1 Determination of biofilm distribution throughout the bioreactor
3.8.1.1 Circumferential analysis
Samples were obtained around the circumference of the reactor. Four samples
were necessary; each located 90° from each other. Refer to Figure 3.4 for sampling
strategy. Samples were obtained from the top (4 cm), middle (8 cm from top of reactor),
and 4 cm from the bottom of the reactor (Figure 3.5). This sampling strategy was used to
minimize errors and provide a better understanding of the effect of hydrodynamics on
biofilm distribution. A total of 12 samples were obtained from the three cross-sectional
areas. The longitudinal and circumferential sampling of biofilm would give an estimate
of the whole area and distribution covered by the biofilm.
3.8.1.2 Longitudinal Analysis
The biofilm growth in the longitudinal direction of the bioreactor was determined
to understand the biofilm distribution throughout the HFMBR. The flow characteristics
were determined by analyzing biofilm growth (mass per membrane surface area) in the
longitudinal direction. The reactor membranes have an approximate length of 20 cm;
therefore, five different samples were taken from different locations identified in Figure
3.4 (bacteria mass and volume).
36
Figure 3.5 Different cross-sections within the HFMBR
The biofilm thickness was also determined by extracting and freezing membranes
from the longitudinal direction. The membranes were analyzed under a Watec CCD
Camera (Edmund Optics VZM450 zoom lense) and a NI IMAQ ((PCI 1200 interface) for
image capture. The thickness was used to determine the density of biofilm present.
3.8.1.3 Outside-in Strategy
Taking in consideration the 3600 membranes within the reactor, samples were
taken from the outside towards the middle of the reactor. The initial idea was to samples
in bundles, and sample from the outside towards the center of the membranes. However,
no biofilm growth was observed in the inner membranes, within a distance of 0.5 cm
from the outer membranes. Due to the size and distribution of the membranes just one
sampling within the center of the reactor was performed. The membrane sampling can be
observed in Figure 3.6.
37
Figure 3.6 Membrane sampling
The last sampling strategy was used to estimate the effect of membrane locations
on biofilm density. Wet and dry masses, as well as the biofilm thickness were measured
to try to estimate biofilm density. A total suspended solids test was conducted following
the procedure in Standard Methods (APHA, 1998). Previously it has been observed that
hydrodynamics affects biofilm density. Assuming full flow contact through all the
membranes, a denser biofilm was assumed to be found within the inner membranes of the
HFMBR, where they are closely packed in comparison to the outside membranes.
3.8.2 Identification of the Bacteria Present within the Biofilm
The purpose of identifying the bacteria present within the bioreactor was to
distinguish between the autotrophic population (Nitrosomonas and Nitrobacter species)
and the possible presence of a heterotrophic population in the nitrifying reactor. In order
to identify the bacteria present within, the biofilm three different plating methods were
applied.
38
3.8.2.1 Identification of Heterotrophic Bacteria by Cultivation in Nutrient Agar
The HFMBR was fed with inorganic carbon during its operation. The presence of
an autotrophic bacteria population was therefore established by feeding inorganic carbon
into the system. However, heterotrophs are found everywhere, and their possible
presence was assumed. Thus, in order to identify the presence of heterotrophic bacteria
within the HFMBR a Nutrient Agar was selected for the identification of a wide variety
of microorganisms. The Nutrient Agar composition is presented in Table 3.3.
Table 3.3 Nutrient Agar Composition Ingredient Amount [g/L]
Beef Extract 3 Peptone 5
Agar 15 Preparation of Medium: Suspend 23 g of the powder in 1 L of purified water. Mix thoroughly. Heat with frequent agitation and boil for 1 min. to completely dissolve the powder. Autoclave at 121° C for 15 min.
3.8.2.2 Identification of Nitrobacter Species by AT5N Medium An autotrophic population was established by feeding inorganic carbon into the
system. In order to identify the presence of autotrophs within the system two different
mediums were taken in consideration, an AT5N Medium and the Winogradsy’s Medium,
Modified. The AT5N Medium was chosen over the Winogradsy’s Medium, Modified due
to the fact that suspended precipitates exist in the latter. The AT5N Medium was
therefore chosen for cultivation and maintenance of the Nitrobacter species and
Nitrobacter Winogradsky, respectively. Table 3.4 presents a detailed composition of the
39
Medium used. For the identification of the Nitrobacter species the original medium was
altered, as for the amount of CaCO3 (due to the hardness of the water in Lubbock) and the
agar added and the replacement of (NH4)2SO4 for an appropriate nitrite source, sodium
nitrite (NaNO2), respectively.
Table 3.4 AT5N medium composition Ingredient Amount [g/L]
CaCl22H2O 0.02 Preparation of Medium: Add components to tap water and bring volume to 1.0 L. Mix thoroughly. Gently heat and bring to boiling. Distribute into tubes or flasks. Autoclave for 15min at 15 psi pressure 121° C.
3.8.2.3 Identification of Nitrosomonas Europaea by AT5N Medium The presence of the Nitrosomonas Europaea was determined by using the AT5N
Medium as well. The Nitrosomonas bacteria are in charge of the oxidation of +4NH -N to
−2NO -N, therefore is the first step for nitrification to reach completion. Thus, analysis for
the identification of these bacteria was accomplished by utilizing the AT5N Medium. The
medium was altered by addition of agar (15 g/L) for solidification and plating purposes
(Table 3.5).
40
Table 3.5 AT5N Medium Composition Ingredient Amount [g/L]
CaCO3 1 (NH4)2SO4 1.5
K2HPO4 0.5 MgSO4 0.05 KHCO3 0.03
CaCl22H2O 0.02 Preparation of Medium: Add components to tap water and bring volume to 1.0 L. Mix thoroughly. Gently heat and bring to boiling. Distribute into tubes for flasks. Autoclave for 15min at 15 psi and 121° C.
Three different plating media were therefore used for identification of autotrophic
and heterotrophic bacteria population. Two replicates of each media at two different
dilutions at the different locations (Figure 3.4) were considered to minimize errors. The
two different dilutions, 1:1 and 1:10, respectively, were used to simulate nutrient rich and
nutrient poor environments to obtain better determination of the presence and possible
bacterial count. Plates were incubated under 30°C for a period of two to seven days
depending on bacterial growth.
41
CHAPTER IV
RESULTS AND DISCUSSION
The objective of this research was to determine and compare the advantages and
disadvantages of a microporous hollow fiber membrane bioreactor (HFMBR) to silicone
membranes. The following discussion details the results obtained for the analysis of
hydrodynamics, mass transfer, biofilm distribution throughout the bioreactor, and the
overall bioreactor performance.
4.1 Nitrification
The change in nitrification capacity of the hollow fiber membrane bioreactor
(HFMBR) is observed in Figure 4.1. The influent +4NH -N concentrations are slightly
variable throughout the experiments due to variations in feed make up, which affect the
standard deviation and results obtained in this report. Conversely, a constant pattern,
where the effluent +4NH -N decreases as the −
3NO -N increases is observed, and relatively
constant −2NO -N concentrations were found throughout the operation of the bioreactor.
Greater concentrations of −3NO -N (mg/L) compared to −
2NO -N were found in the
effluent. This fact indicates that complete nitrification ( +4NH → −
2NO → −3NO )
conversion was achieved. An HRT of 0.15 days, showed the greatest +4NH -N conversion
to −3NO -N, which is attributed to the fact that greater nitrification is obtained at longer
From Figure 4.16, it can be observed that a thicker biofilm was formed at the
bottom of the reactor, which is as expected due to the higher concentrations of nutrients
closer to the influent port of the HFMBR. Figures 4.17 and 4.18 show some of the
thicknesses obtained when analyzing the membranes under a Watec CCD Camera.
79
It was noted that the uneven distribution of biofilm was due to flow
maldistribution. The influent feed was pumped in the reactor from the bottom of the
bioreactor (Figures 3.1, 3.3); therefore, the flow was unequally distributed through the
HFMBR. Biofilm formation was observed in the same direction of the moving flow;
therefore, the assumption of an existing preferential flow path was taken in consideration
for this analysis. Channeling due to biofilm formation on the membranes was also
assumed to have an effect on the hydrodynamic behavior within the bioreactor.
Biofilm thickness was also found to be greater at the bottom of the reactor, which
supports the theory that the biofilm is affected by nutrient consumption. Nutrient
concentrations decrease as the fluid passes through the reactor; therefore, it can be
assumed that thinner biofilms would be found along the fluid’s path. However, this was
not the case shown in Figure 4.16; it was attributed primarily to the assumption of
uniform thickness within a sampling section (1 cm) and due to the possible sampling
errors when performing the experimental analysis.
Figures 4.17 and 4.18 present the different biofilm thickness at different locations
within the HFMBR. However, as expected, more biofilm grew around the bottom of the
reactor. These results suggest that the shear forces due to velocity affected biofilm
structure (Casey et al., 2000). Nevertheless, these results coincide with Gibbs and Bishop
(1995) who state that the velocity of the bulk fluid flow is one factor affecting the biofilm
thickness and density where higher velocities result in compressed (or thinner)
concentration of the biofilm. Biofilm density is not presented in this thesis due to the
sampling methods used and high error probability.
80
Figure 4.17 Biofilm thickness of sections 2A and 3A
81
Figure 4.18 Biofilm thickness of sections 4A and 5A
82
The purpose of determining the biofilm distribution throughout the bioreactor was
to identify the preferential flow path, and to predict the effect of biofilm formation on
hydrodynamics. Due to the high number of membranes and the packing density, flow did
not pass by all the membranes; the fluid flow was found to contact the outer membranes,
like in an annulus. The hydrodynamics were meant to be understood by analyzing the
biofilm growth; however, no definite conclusion was obtained. The biofilm was assumed
to affect the flow pattern, but it could be concluded that the flow pattern also affected the
biofilm growth within the HFMBR. Results have shown that the HFMBR did not operate
at maximum capacity. The loading rates, mass transfer, and hydrodynamics have been
affected by the bioreactor configuration; therefore, not operating at maximum efficiency.
4.5.5 Bacteria Identification
The reactor feed was composed of inorganic carbon in order to support an
autrotrophic population. Due to biofilm structure and heterogeneity, the possible
presence of a heterotrophic population was assumed. Samples were extracted from four
different locations at three cross-sectional areas, at the top, middle, and bottom of the
bioreactor (Figures 4.11 and 3.5). Two different medias were used to identify the
bacteria within the HFMBR, a Nutrient Agar and a AT5N Medium, for identification of
heterotrophic and autotrophic microorganisms, respectively.
Bacterial plates were made at two different dilutions (1:10 and 1:1) in order to
have better identification. Plates were incubated for 5 days; however rapid growth was
observed on the nutrient plates (heterotrophs) after 2 days while no growth was observed
83
for the nitrosomonas and nitrobacter species. As expected, heterotrophs grew faster than
autotrophs; however, the plates were incubated for the same period of time (5 days).
For all sections, heterotrophs, nitrosomonas, and nitrobacter species were found
to exist within the HFMBR. Heterotrophs were found even though the feed consisted of
inorganic carbon. The presence of heterotrophs was attributed to the initial feed used,
which contained of urine. The existence and survival of the heterotrophic population
within the HFMBR can be explained by the fact that these microorganisms fed on the
dead autotrophic bacterial cells. However, the fact that more colonies were observed
within the plates was considered to be a function of incubation time. Greater growth
occurred due to a larger incubation period of time instead of a larger population existing
within the bioreactor. Although, plate counts were beyond the scope of this thesis, a
greater population of nitrosomonas species was expected in comparison to the
nitrobacter specie. As previously mentioned, the feed consisted of inorganic carbon
(NH4Cl), which is the energy and carbon source for this specie, as opposed to −2NO -N for
nitrobacter specie.
84
CHAPTER V
CONCLUSIONS
A commercial microporous hollow fiber bioreactor (HFMBR) was analyzed for
wastewater treatment applications. The goal of this research was to (a) evaluate the
effect of mass transfer by the use of microporous membranes; (b) compare the effect of
membrane type and configuration on treatment efficiency to previous literature values;
and (c) determine the amount and distribution of biofilm growth within the reactor.
Results of this study, if appropriate, will help determine the use of the HFMBR in space
applications.
Loading rates and nitrification studies were performed in order to determine the
reactor’s optimum performance. High nitrification efficiencies can be obtained due to the
high surface area presented by the use of the microporous membranes in a HFMBR. In
this thesis, due to the high packing density and configuration of the bioreactor, the total
membrane surface area was not utilized and the reactor did not operate at its maximum
removal efficiency. However, for aeration purposes, the HFMBR may be appropriate for
wastewater treatment if the configuration is arranged to where the packing density is
reduced by increasing the spacing between membranes. Therefore, increasing the surface
area available for biofilm growth in the reactor would increase the removal efficiency,
being more suitable for wastewater treatment. If membrane spacing is not improved an
increase in reactor volume would be another option to consider for wastewater treatment.
However, for space applications, this would not be a recommended option due to
85
NASA’s requirements to reduce mass and volume of treatment systems owing to space
limitations.
Mass transfer within the HFMBR would also aid in the determination of the
design and overall reactor’s performance. The microporous membranes within the
HFMBR were considered to provide a negligible mass transfer resistance; thus, the
HFMBR has been found to provide better mass transfer in comparison to a membrane
system for which the membrane resistance is taken in consideration. Bubble-less aeration
was not achieved during the initial experiments without biofilm; however, bubbles were
not observed after biofilm formation. Mass transfer improved after biofilm formation
due to higher oxygen flux diffusing in to the biofilm increasing the total flux or mass
transfer of the bulk liquid.
The correlations between the oxygen mass transfer and fluid dispersion were
considered for the reactor performance. The bioreactor hydrodynamics were affected by
the reactor configuration, the number of membranes, and flow velocity. The effect of
flow velocity on reactor performance was mainly of interest for mass transfer analysis.
Flow velocity influenced mass transfer in the diffusion boundary layer thickness at the
biofilm liquid interface and biofilm density. Convective flow was found to be the
controlling factor on the rate of mass transfer due to the reactor size and packing density.
Uneven biofilm distribution was also found to affect the hydrodynamics of the bioreactor
due to channeling; thus, fluid flowed through preferential paths in regions of lower
packing, reducing the working volume and overall performance of the bioreactor.
86
In the HFMBR biofilm formation, thickness and density were proven to be
dependent on the fluid velocity and bioreactor configuration for which, uneven mass,
density, and thickness were observed. The morphological characteristics of biofilms such
as thickness, density, and shape are very important for the overall performance of a
biofilm reactor. These characteristics strongly affect the mass transfer in a biofilm reactor
and overall reactor performance. The biofilm heterogeneity also played an important role
in the reactor performance. The presence of a heterotrophic and autotrophic (nitrifiers)
was identified. Maximum growth was achieved by the nitrifiers as a maximum 75%
conversion rate was achieved. Maximum growth rate would not vary; however, higher
conversion rates would be achieved by increasing the nitrifying population, which will
increase conversion rates and bioreactor performance.
The advantages and disadvantages of using parallel distributed microporous
membranes within the HFMBR were analyzed. Microporous membranes in a HFMBR
present the advantage of increasing oxygen transport through the membranes in
comparison to silicone membranes. The use of parallel membranes at high packing
densities might not enhance the mass transfer in comparison to random membranes due
to biofilm formation and channeling. Short circuiting caused by channeling did occur
within the HFMBR; nevertheless, microporous membranes present higher surface area
for biofilm formation, presenting a better choice for high mass transfer to occur and high
ammonia removal rates.
The HFMBR is capable of achieving maximum nitrification efficiency at low
loading rates and high HRTs. Due to the reactor’s size, the HFMBR presents promising
87
use in space applications. These characteristics offer a potential for NASA’s needs;
nonetheless, developing a system with more favorable system hydrodynamics would aid
to improve treatment efficiency in a HFMBR.
Considering the above, it is recommended that to adequately demonstrate the
effectiveness of HFMBRs for wastewater applications under microgravity conditions a
comparison of the performance of several units is still needed. Research on the effect of
different membrane models, analyzing membrane geometry (parallel vs. random
membranes) should be addressed. Also, the determination of best operating pressure
without bubble formation, and the effect of increasing temperature for the enhancement
of treatment are factors to consider.
.
88
REFERENCES
Ahmed, T. and Semmens, M.J. 1992. The use of independently sealed microporous hollow fiber membranes for oxygenation of water: model development, J. Membrane Sci., 69, 11-20.
Ahmed, T. and Semmens, M.J. 1992. Use of sealed end hollow fibers for bubbleless
membrane aeration: experimental studies. J. Membrane Sci., 69, 1-10. Alphenaar P. A., Perez C. and Lettinga C. 1993. The influence of substrate transport on
porosity and methanogenic activity of anaerobic sludge granules. Appl. Microb. Biothech., 39, 276-280
APHA, AWWA, and WEF. 1998. Standard Methods for the Examination of Water and
Wastewater, 20th Edition. Atlas, R.M. 1995. Handbook of Media for Environmental Microbiology. CRC Press,
Boca Raton, FL. Bao, L. and Lipscomb, G. 2002. Well-developed mass transfer in axial flows through
randomly packed fiber bundles with constant wall flux. Chem. Eng. Sci., 57, 125-132. Bao, L. and Lipscomb, G. 2002. Mass transfer in axial flows through randomly packed
fiber bundles with constant wall concentration. J. Membrane Sci., 204, 207-220. Bishop, P.L., Zhang, T.C., Fu, Y.C. 1995. Effects of biofilm structure, microbial
distributions and mass-transport on biodegradation processes. Wat. Sci. Technol., 31(1), 143-152.
Bovendeur, J., Zwage, A. B., Lobee, B. and Blom, J.H. 1990. Fixed-film reactors on
aquaculture water recycle system: Effect of organic matter elimination on nitrification kinetic. Wat. Res., 24(2), 207.
Brindle, K. and Stephenson, T. 1996. Nitrification in a bubbleless oxygen mass transfer
membrane bioreactor. Wat. Sci. Technol., 34(9), 261-267. Brito A. G. and Melo L.F. 1999. Mass transfer coefficients within anaerobic biofilms:
effects of external liquid velocity. Wat. Res., 33(17), 3673-3678.
89
Casey E., Glennon B., Hamer, G. 1999. Biofilm development in a membrane aerated biofilm reactor: effect of flow velocity on performance. Biotechnol. Bioeng., 67(4) 446-486.
Casey E., Glennon B., and Hamer, G. 1999. Review of membrane aerated biofilm
reactors. Resources, Conservation and Recycling (27), 203-215. Casey, E., Glennon, B., Hamer, G. 2000. Biofilm development in a membrane-aerated
biofilm reactor: Effect of flow velocity on performance. Biotechnol. Bioeng., 67(4), 476-86.
removal in wastewater using phosphorylated PVA-immobilized microorganisms. Enzyme Microb. Technol., 23, 311-320.
Cote, P. 1989. Bubble-free aeration using membranes: Mass transfer analysis. J.
Membrane Sci., 47, 91-106. Cussler E. L. 2002. Diffusion, mass transfer in fluid systems. 2nd Eddition. Cambridge
University Press, U.K. Davis, D. A., and Arnold, C. R. 1998. The design, Management and production of a
recirculating raceway system for the production of marine shrimp. Aquacultural Eng., 17(3), 193.
Finger, B.W., Supra, L.N., DallBauman, L., Pickering, K.D. 1999. Development and
Testing of Membrane Biological Wastewater Processor, Proceedings 29th International Conference on Environmental Systems (ICES).
Garrido, J.M., van Benthum, W.A.J., van Loosdrecht, M.C.M. and Heijnen, J.J. 1997.
Influence of dissolved oxygen concentration on nitrite accumulation in a biofilm airlift suspension reactor. Biotechnol. Bioeng., 53(2), 168–178.
Geankoplis J. Christie. 1983. Transport processes: Momentum, Heat, and Mass. Allyn
and Bacon, Inc. Newton, MA. Gibbs, J.T. and Bishop, P.L. 1995. A method for describing biofilm surface roughtness
using geostatistical techniques. Wat. Sci. Tech., 32 (8), 91-98. Greiner, A., and Timmons, M. 1998. Evaluation of the nitrification rates of microbead
and trickling filters in an intensive recirculation tilapia production facility. Aquacultural Eng., 18(3)189.
90
Grimberg, S. J., Rury, M. J., Jimenez, K. M., and Zander, A. K. 2000. Trinitrophenol treatment in a hollow fiber membrane biofilm reactor. Wat. Sci. Technol., 41 (4), 235-238.
Hibiya, K., Terada, A., Tsuneda, S., and Hirata, A. 2003. Simultaneous nitrification and denitrification by controlling vertical and horizontal microenvironment in a membrane aerated biofilm reactor. J. Biotech., 100, 23-32.
Jackson, A., Morse, A., Ho, T. and Collins, G. 2004. Optimization and Biochemical
Characterization of Biological Reactors for Long-Term Space Habitation Waste Streams, 34th International Conference on Environmental Systems (ICES), Doubletree Hotel, Colorado Springs, CO, July 19-22.
Kamstra, A., Heul, J. W.and Nihof, M. 1998. Performance and optimization of trickling
filters on eel farms. Aquacultural Eng., 17(3), 175. Knudsen, J.G. and Katz, D.L. 1958. Fluid dynamics and heat transfer. McGraw-Hill,
New York, NY. Kreulen , H. Smolders, C.A, Versteeg, G.F., and Van Swaaij W.P.M. 1993. Microporous
hollow fibre membrane modules as gas-liquid contactors. Part 1. Physical mass transfer processes. A specific application: mass transfer in highly viscous liquids. J. Membrane Sci., 78, 197-216.
Kuenen, J.G., Robertson, L.A. 1988. Ecology of nitrification and denitrification. In: Cole,
J.A., Ferguson, S.J. (Eds.), The Nitrogen and Sulphur Cycles. Cambridge University Press, Cambridge, pp. 161-218.
Lewandowski, Z., Stoodley, P., Altobelli, S. 1995. Experimental and conceptual studies
on mass-transport in biofilms. Water Sci. Technol., 31(1) 153-162. Maier, Raina M., Pepper, Ian L., and Gerba, Charles P. 2000. Environmental
Microbiology. Academic Press, San Diego, CA. Marshall, K. C., 1985. Mechanisms of bacterial adhesion at solid-liquid interfaces. In
Bacterial Adhesion: Mechanisms and Physiological Significance. D.C. Savage, and M. Fletcher, eds. Plenum, NY, pp 131-161.
McLamore, E., Morse, A., Jackson, A. 2004. Incorporation of a Membrane-Aerated
Bioreactor in a Water Recovery System, 34th International Conference on Environmental Systems (ICES), Doubletree Hotel, Colorado Springs, Colorado, July 19-22.
91
Menoud, P., Barton, G.W. 1999. Simultaneous nitrification and denitrification using siporax ™ packing. Wat. Sci. Technol., 40 (4-5), 153-160.
Metcalf & Eddy, Inc. 2003. Wastewater engineering treatment and reuse. 4th Edition.
McGraw-Hill, NY. Morse, A., Jackson, A., Rainwater, K. and Pickering, K. 2003. Nitrification using a
Membrane-Aerated Biological Reactor, 33rd International Conference on Environmental Systems (ICES), Bayshore Resort & Marina, Vancouver, B.C., Canada, July 7-10.
Muirhead, D., Rector, T., Collins, G., Keister, H., Morse, A., Jackson, A., Pickering, K.
2003. Performance of a Small-Scale Advanced Life Support Water Recovery System, 33rd International Conference on Environmental Systems (ICES), Bayshore Resort & Marina, Vancouver, B.C., Canada, July 7-10.
Nijhof, M. and Bovendeur, J. 1990. Fixed film nitrification characteristics in sea-water
recirculation fish culture systems. Aquaculture, 87, 133. Petersen, J. N. and Davison, B. H. 1991. Modeling of an Immobilized Cell, Three-Phase
Fluidized-Bed Bioreactor. Appl. Biochem. Biotech. 28 (29), 685-98. Pickering, K.P., and Edeen M.A. 1997. Lunar-Mars life support program phase III water
recovery system operation and results. Society of Automotive Engineers Conference. Paper #981707.
Rittmann, B. and McCarty, P.L. 2001. Environmental Biotechnology: Principles and
Applications. The McGraw-Hill Companies, Inc. New York, NY. Rittmann, B. E.; Regan J. M.; and Stahl D. A. 1994. Nitrification as a source of organic
substrate in biological treatment. Wat. Sci. Technol., 30(6), 1-8. Semmens, M.J., and Essila, N.J. 2001. Modeling biofilms on gas-permeable supports:
Strigle, Ralph F. 1987. Random packings and packed towers. Design and Applications. Gulf Publishing Co. Houston, TX.
Suzuki, Y., Hatano, N., and Ikeda, H. 2000. Performance of nitrogen removal and biofilm
structure of porous gas permeable membrane reactor. Wat. Sci. Technol., 41(4), 211-217.
92
Tavares, C. R. G; Araujo Rocha, R. R; and Guedes, T. A. 2001. Nitrification and Denitrification Processes for Biologic Treatment of Industrial Effluents. App. Biochem. Biotechnol., 91(93), 437-446.
Tijhuis, L., Gjaltema A, van Loosdrecht M.C.M & Heijnen J.J. 1995. Detachment of
biomass from suspended nongrowing spherical biofilms in airlift reactors. Biotechnol. Bioeng. 46, 258–2.
Timberlake, Dennis L., Strand Stuart E., and Williamson, K. J. 1988. Combined aerobic
heterotrophic oxidation, nitrification and denitrification in a permeable-support biofilm. Wat. Res. 22 (12), 1513-1517.
Torrey, S. 1984. Membrane and ultrafiltration technology: developments since 1981.
Noyes Data Corp. Park Ridge, N.J. Udert, K.M., Fux, C., Munster, M., Larsen, T.A., Siegrist, H. and Gujer, W. 2003.
Nitrification and autotrophic denitrification of source-separated urine. Wat. Sci. Technol. 48(1), 119-130.
Van Rijn, J and Rivera, G. 1990. Aerobic and anaerobic biofiltration in an aquaculture
unit-nitrite accumulation as a result of nitrification and denitrification. Aquacultural Eng. 9, 217
Viera, M. J., Melo, L., Pinheiro, M. M. 1993. Biofilm formation: Hydrodynamic effects
on internal diffusion and structure. Biofouling 7, 67-80. Watson, S.W., F.W. Valos, and Waterbury, J.B. 1981. The Family Nitrobacteraceae. In
The Prokaryotes. M.P. Starr et al. Berlin: Springer-Verlag Wickins, J. F. 1985. Ammonia production and oxidation during the culture of marine
prawns and lobsters in laboratory recirculation systems. Aquacultural Eng. 4, 155. Wickramasinghe, S.R., Semmens, M.J., and Cussler, E.L. 1992. Mass transfer in various
hollow fiber geometries. J. Membrane Sci., 69, 235-250. Wickramasinghe, S.R. and Han, Binding. 2002. Mass and momentum transfer in
commercial blood oxygenators. Desalination, 148, 227-233. Yamagiwa, K. and Ohkawa A. 1994. Simultaneous organic carbon removal and
nitrification by biofilm formed on oxygen enrichment membrane. J. Chem. Eng. of Japan. 27(5) 638-643.
Yang, M., and Cussler, E.L. 1986. Designing Hollow-fiber contactors. AIChE Journal. 32
(11), 1910-1916.
93
Zhang, T. C., Bishop, P.L. 1994. Evaluation of tortuosity factors and effective
diffusivities in biofilms. Wat. Res. 28(11) 2279-2287. Zhu, S. and Chen, S. 1999. An experimental study on nitrification biofilm performances
using a series reactor system. Aquacultural Eng. 20(4), 245.
94
APPENDIX A
RAW DATA DURING THE HFMBR STUDY
95
Table A1. Test point 1. Influent Concentrations (mg/L)
Figure B2. Raw data for dissolved oxygen and ammonia
117
HFMDays
0 50 100 150 200In
f. N
Ox
(mg/
L)
0
5
10
15
20
25
30
Days
0 50 100 150 200
Effl.
NO
x (m
g/L)
0
20
40
60
80
100
120
140
160
x =2.37std = 1.1
x = 87.33std = 9.32
x = 3.22std = 2.23
x = 62.26std =10.7
x = 2.93std = 1.13
x = 42.51std = 7.20
x = 2.49std = 1.58
x = 61.76std = 7.43
x = 4.03std = 2.23
x = 65.71std = 7.84
Flow Rate 0.5 mL/min
Flow Rate1.0 mL/min
Flow Rate 1.5 mL/min
Flow Rate0.6 mL/min
Flow Rate 0.7 mL/min
Flow Rate0.3 mL/min
x = 5.84std = 4.35
x = 100.94std = 16.0
Flow Rate1.7 mL/min
x = 2.15std = 0.54
x = 21.65std = 5.53
Figure B3. Raw data for nitrogen oxides
APPENDIX C
MASS TRANSFER CALCULATIONS
118
INIT
IAL
DO
CA
LCU
LATI
ON
S
Flux
Cal
cula
tions
V=
0.06
3 L
Q
= 0.
3 cm
3 /min
Q=
cm
3 /min
Ope
ratin
g Pr
essu
re=
0.5
psi
O
pera
ting
Pres
sure
= 0.
0340
2286
at
m
H
enry
's co
nsta
nt fo
r O2=
41
100
atm
Surf
ace
Are
a pe
r HF=
1.
4645
cm
2
Tota
l Mem
bran
e Su
rfac
e A
rea=
52
72.2
cm
2
CL
(0)=
0.
45
mg-
DO
/L
M
embr
ane
Thic
knes
s=
0.00
8 cm
Mem
bran
e Th
ickn
ess=
0.
0000
8 m
Ta
ble
C1.
Initi
al D
O c
alcu
latio
ns a
t 0.3
mL/
min
Tim
e C
L (ti
me)
C
hang
e in
Con
c C
hang
e in
Ti
me
Flux
(J)
Flux
(J)
Flux
(J)
Cha
nge
in C
onc
ko
[min
] [m
g-D
O/L
] [m
mol
-DO
] [m
in]
[mm
ol-D
O/m
in-c
m2 ]
[mm
ol-D
O/m
in-m
2 ] [m
g-D
O/m
in-
m2 ]
[mg-
DO
/m3 ]
[m/m
in]
0 0.
45
- -
- -
- -
- 10
2.
18
3.41
E-03
1.
00E+
01
6.46
E-08
6.
46E-
04
2.07
E-02
1.
73E+
03
1.19
E-05
20
2.
35
3.35
E-04
1.
00E+
01
6.35
E-09
6.
35E-
05
2.03
E-03
1.
70E+
02
1.19
E-05
25
2.
37
3.94
E-05
5.
00E+
00
1.49
E-09
1.
49E-
05
4.78
E-04
2.
00E+
01
2.39
E-05
35
2.
92
1.08
E-03
1.
00E+
01
2.05
E-08
2.
05E-
04
6.57
E-03
5.
50E+
02
1.19
E-05
55
3.
26
6.69
E-04
2.
00E+
01
6.35
E-09
6.
35E-
05
2.03
E-03
3.
40E+
02
5.97
E-06
85
3.
78
1.02
E-03
3.
00E+
01
6.47
E-09
6.
47E-
05
2.07
E-03
5.
20E+
02
3.98
E-06
10
0 3.
82
7.88
E-05
1.
50E+
01
9.96
E-10
9.
96E-
06
3.19
E-04
4.
00E+
01
7.97
E-06
14
0 3.
85
5.91
E-05
4.
00E+
01
2.80
E-10
2.
80E-
06
8.96
E-05
3.
00E+
01
2.99
E-06
45
5 3.
89
7.88
E-05
3.
15E+
02
4.74
E-11
4.
74E-
07
1.52
E-05
4.
00E+
01
3.79
E-07
119
Tabl
e C
2. In
itial
DO
cal
cula
tions
at 5
mL/
min
Tim
e C
L (ti
me)
C
hang
e in
Con
c
Cha
nge
in
Tim
e Fl
ux (J
) Fl
ux (J
) Fl
ux (J
) C
hang
e in
C
onc
ko
[min
] [m
g-D
O/L
] [m
mol
-DO
] [m
in]
[mm
ol-D
O/m
in-c
m2 ]
[mm
ol-D
O/m
in-m
2 ] [m
g-D
O/m
in-m
2 ] [m
g-D
O/m
3 ] [m
/min
] 0
0.67
-
- -
- -
- -
5 2.
52
3.64
E-03
5
1.38
E-07
1.
38E-
03
4.42
E-02
18
50
2.39
E-05
10
2.
66
2.76
E-04
5
1.05
E-08
1.
05E-
04
3.35
E-03
14
0 2.
39E-
05
15
2.79
2.
56E-
04
5 9.
71E-
09
9.71
E-05
3.
11E-
03
130
2.39
E-05
30
2.
85
1.18
E-04
15
1.
49E-
09
1.49
E-05
4.
78E-
04
60
7.97
E-06
55
2.
86
1.97
E-05
25
1.
49E-
10
1.49
E-06
4.
78E-
05
10
4.78
E-06
65
2.
89
5.91
E-05
10
1.
12E-
09
1.12
E-05
3.
58E-
04
30
1.19
E-05
15
0 2.
89
0.00
E+00
85
0.
00E+
00
0.00
E+00
0.
00E+
00
0 #D
IV/0
! 17
0 2.
89
0.00
E+00
20
0.
00E+
00
0.00
E+00
0.
00E+
00
0 #D
IV/0
!
120
Tabl
e C
3. In
itial
DO
cal
cula
tions
at 1
0 m
L/m
in
Tim
e C
L (ti
me)
C
hang
e in
Con
c
Cha
nge
in
Tim
e Fl
ux (J
) Fl
ux (J
) Fl
ux (J
) C
hang
e in
C
onc
ko
[min
] [m
g-D
O/L
] [m
mol
-DO
] [m
in]
[mm
ol-D
O/m
in-
cm2 ]
[mm
ol-D
O/m
in-
m2 ]
[mg-
DO
/min
-m2 ]
[mg-
DO
/m3 ]
[m/m
in]
0 0.
45
- -
- -
- -
-
10
2.53
4.
10E-
03
10
7.77
E-08
7.
77E-
04
2.49
E-02
20
80
1.19
E-05
15
2.56
5.
91E-
05
5 2.
24E-
09
2.24
E-05
7.
17E-
04
30
2.39
E-05
30
2.58
3.
94E-
05
15
4.98
E-10
4.
98E-
06
1.59
E-04
20
7.
97E-
06
40
2.6
3.94
E-05
10
7.
47E-
10
7.47
E-06
2.
39E-
04
20
1.19
E-05
55
2.65
9.
84E-
05
15
1.24
E-09
1.
24E-
05
3.98
E-04
50
7.
97E-
06
70
2.66
1.
97E-
05
15
2.49
E-10
2.
49E-
06
7.97
E-05
10
7.
97E-
06
100
2.66
0.
00E+
00
30
0.00
E+00
0.
00E+
00
0.00
E+00
0
#DIV
/0!
145
2.66
0.
00E+
00
45
0.00
E+00
0.
00E+
00
0.00
E+00
0
#DIV
/0!
160
2.67
1.
97E-
05
15
2.49
E-10
2.
49E-
06
7.97
E-05
10
7.
97E-
06
121
Ta
ble
C4.
Initi
al D
O c
alcu
latio
ns a
t 21
mL/
min
Tim
e C
L (ti
me)
C
hang
e in
Con
c
Cha
nge
in
Tim
e Fl
ux (J
) Fl
ux (J
) Fl
ux (J
) C
hang
e in
Con
c ko
[min
] [m
g-D
O/L
] [m
mol
-DO
] [m
in]
[mm
ol-D
O/m
in-c
m2 ]
[mm
ol-D
O/m
in-m
2 ] [m
g-D
O/m
in-m
2 ] [m
g-D
O/m
3 ] [m
/min
] 0
0.9
- -
- -
- -
- 8
3.05
4.
23E-
03
8 1.
00E-
07
1.00
E-03
3.
21E-
02
2150
1.
49E-
05
10
3.07
3.
94E-
05
2 3.
73E-
09
3.73
E-05
1.
19E-
03
20
5.97
E-05
13
3.
08
1.97
E-05
3
1.24
E-09
1.
24E-
05
3.98
E-04
10
3.
98E-
05
14
3.18
1.
97E-
04
1 3.
73E-
08
3.73
E-04
1.
19E-
02
100
1.19
E-04
18
3.
21
5.91
E-05
4
2.80
E-09
2.
80E-
05
8.96
E-04
30
2.
99E-
05
28
3.3
1.77
E-04
10
3.
36E-
09
3.36
E-05
1.
08E-
03
90
1.19
E-05
48
2.
84
-9.0
6E-0
4 20
-8
.59E
-09
-8.5
9E-0
5 -2
.75E
-03
-460
5.
97E-
06
58
2.76
-1
.58E
-04
10
-2.9
9E-0
9 -2
.99E
-05
-9.5
6E-0
4 -8
0 1.
19E-
05
68
2.31
-8
.86E
-04
10
-1.6
8E-0
8 -1
.68E
-04
-5.3
8E-0
3 -4
50
1.19
E-05
73
2.
31
0.00
E+00
5
0.00
E+00
0.
00E+
00
0.00
E+00
0
#DIV
/0!
113
2.36
9.
84E-
05
40
4.67
E-10
4.
67E-
06
1.49
E-04
50
2.
99E-
06
122
Tabl
e C
5. In
itial
DO
cal
cula
tions
. Sum
mar
y ta
ble
Flow
rate
A
ir Pr
essu
re
Max
imum
Flu
x ko
(ove
rall)
1/
ko (o
vera
ll)
1/kL
(liq
uid)
Sh
D
iffus
ion
[mL/
min
] [p
si]
[mm
ol-D
O/m
2 -min
] [m
/min
] [m
in/m
] [m
in/m
] [-
] [m
2 /min
]
0.3
0.5
0.00
0646
018
1.19
E-05
83
685.
71
8368
5.71
20
1.
18E-
07
5 0.
5 1.
1202
6E-0
5 1.
1949
5E-0
5 83
685.
71
8368
5.71
20
1.
18E-
07
10
0.5
0.00
0776
716
1.19
495E
-05
8368
5.71
83
685.
71
20
1.18
E-07
21
0.5
0.00
1003
569
1.49
368E
-05
6694
8.57
66
948.
57
25
1.18
E-07
123
FIN
AL
DO
CA
LCU
LATI
ON
S
Fl
ux C
alcu
latio
ns
V=
0.06
3 L
Q=
0.3
cm3 /m
in
Q=
cm
3 /min
O
pera
ting
Pres
sure
= 0.
5 ps
i O
pera
ting
Pres
sure
= 0.
0340
2286
at
m
Hen
ry's
cons
tant
for O
2=
4110
0 at
m
Surf
ace
Are
a pe
r HF=
1.
4645
cm
2 To
tal M
embr
ane
Surf
ace
Are
a=
5272
.2
cm2
CL
(0)=
0
mg-
DO
/L
Mem
bran
e Th
ickn
ess=
0.
008
cm
Mem
bran
e Th
ickn
ess=
0.
0000
8 m
Tabl
e C
6. F
inal
DO
cal
cula
tions
at 0
.3 m
L/m
in
Tim
e C
L (ti
me)
C
hang
e in
Con
c.
Cha
nge
in T
ime
Flux
(J)
Flux
(J)
Flux
(J)
Cha
nge
in C
onc.
ko
[m
in]
[mg-
DO
/L]
[mm
ol-D
O]
[min
] [m
mol
-DO
/min
-cm
2 ] [m
mol
-DO
/min
-m2 ]
[mg-
DO
/min
-m2 ]
[mg-
DO
/m3 ]
[m/m
in]
0 0
- -
- -
- -
- 15
1.
12
2.21
E-03
15
2.
79E-
08
2.79
E-04
8.
92E-
03
1120
7.
97E-
06
50
0.56
-1
.10E
-03
35
-5.9
7E-0
9 -5
.97E
-05
-1.9
1E-0
3 -5
60
3.41
E-06
70
0.
76
3.94
E-04
20
3.
73E-
09
3.73
E-05
1.
19E-
03
200
5.97
E-06
10
5 2.
3 3.
03E-
03
35
1.64
E-08
1.
64E-
04
5.26
E-03
15
40
3.41
E-06
14
0 1.
25
-2.0
7E-0
3 35
-1
.12E
-08
-1.1
2E-0
4 -3
.58E
-03
-105
0 3.
41E-
06
190
2.33
2.
13E-
03
50
8.07
E-09
8.
07E-
05
2.58
E-03
10
80
2.39
E-06
26
5 2.
48
2.95
E-04
75
7.
47E-
10
7.47
E-06
2.
39E-
04
150
1.59
E-06
28
0 2.
6 2.
36E-
04
15
2.99
E-09
2.
99E-
05
9.56
E-04
12
0 7.
97E-
06
124
Ta
ble
C7.
Fin
al D
O c
alcu
latio
ns a
t 1 m
L/m
in
Tim
e C
L (ti
me)
C
hang
e in
C
onc.
C
hang
e in
Tim
e Fl
ux (J
) Fl
ux (J
) Fl
ux (J
) C
hang
e in
Con
c.
ko
[min
] [m
g-D
O/L
] [m
mol
-DO
] [m
in]
[mm
ol-D
O/m
in-
cm2 ]
[mm
ol-D
O/m
in-m
2 ] [m
g-D
O/m
in-
m2 ]
[mg-
DO
/m3 ]
[m/m
in]
0 0.
5 -
- -
- -
- -
15
2.63
4.
19E-
03
15
5.30
E-08
5.
30E-
04
1.70
E-02
21
30
7.97
E-06
20
3.
16
1.04
E-03
5
3.96
E-08
3.
96E-
04
1.27
E-02
53
0 2.
39E-
05
25
2.75
-8
.07E
-04
5 -3
.06E
-08
-3.0
6E-0
4 -9
.80E
-03
-410
2.
39E-
05
40
2.78
5.
91E-
05
15
7.47
E-10
7.
47E-
06
2.39
E-04
30
7.
97E-
06
45
2.61
-3
.35E
-04
5 -1
.27E
-08
-1.2
7E-0
4 -4
.06E
-03
-170
2.
39E-
05
50
2.64
5.
91E-
05
5 2.
24E-
09
2.24
E-05
7.
17E-
04
30
2.39
E-05
60
2.
85
4.13
E-04
10
7.
84E-
09
7.84
E-05
2.
51E-
03
210
1.19
E-05
75
3.
16
6.10
E-04
15
7.
72E-
09
7.72
E-05
2.
47E-
03
310
7.97
E-06
85
3.
17
1.97
E-05
10
3.
73E-
10
3.73
E-06
1.
19E-
04
10
1.19
E-05
10
0 3.
23
1.18
E-04
15
1.
49E-
09
1.49
E-05
4.
78E-
04
60
7.97
E-06
11
0 3.
26
5.91
E-05
10
1.
12E-
09
1.12
E-05
3.
58E-
04
30
1.19
E-05
13
0 3.
29
5.91
E-05
20
5.
60E-
10
5.60
E-06
1.
79E-
04
30
5.97
E-06
13
5 3.
32
5.91
E-05
5
2.24
E-09
2.
24E-
05
7.17
E-04
30
2.
39E-
05
140
3.33
1.
97E-
05
5 7.
47E-
10
7.47
E-06
2.
39E-
04
10
2.39
E-05
15
0 3.
36
5.91
E-05
10
1.
12E-
09
1.12
E-05
3.
58E-
04
30
1.19
E-05
17
0 3.
38
3.94
E-05
20
3.
73E-
10
3.73
E-06
1.
19E-
04
20
5.97
E-06
20
0 3.
41
5.91
E-05
30
3.
73E-
10
3.73
E-06
1.
19E-
04
30
3.98
E-06
21
5 3.
52
2.17
E-04
15
2.
74E-
09
2.74
E-05
8.
76E-
04
110
7.97
E-06
26
0 3.
53
1.97
E-05
45
8.
30E-
11
8.30
E-07
2.
66E-
05
10
2.66
E-06
125
Ta
ble
C8.
Fin
al D
O c
alcu
latio
ns a
t 10
mL/
min
Tim
e C
L (ti
me)
C
hang
e in
C
onc.
C
hang
e in
Tim
e Fl
ux (J
) Fl
ux (J
) Fl
ux (J
) C
hang
e in
Con
c.
ko
[min
] [m
g-D
O/L
] [m
mol
-DO
] [m
in]
[mm
ol-D
O/m
in-c
m2 ]
[mm
ol-D
O/m
in-m
2 ] [m
g-D
O/m
in-
m2 ]
[mg-
DO
/m3 ]
[m/m
in]
0 0
- -
- -
- -
- 5
3.26
6.
42E-
03
5 2.
43E-
07
2.43
E-03
7.
79E-
02
3260
2.
39E-
05
10
2.8
-9.0
6E-0
4 5
-3.4
4E-0
8 -3
.44E
-04
-1.1
0E-0
2 -4
60
2.39
E-05
15
3.
51
1.40
E-03
5
5.30
E-08
5.
30E-
04
1.70
E-02
71
0 2.
39E-
05
20
3.77
5.
12E-
04
5 1.
94E-
08
1.94
E-04
6.
21E-
03
260
2.39
E-05
25
3.
04
-1.4
4E-0
3 5
-5.4
5E-0
8 -5
.45E
-04
-1.7
4E-0
2 -7
30
2.39
E-05
30
3
-7.8
8E-0
5 5
-2.9
9E-0
9 -2
.99E
-05
-9.5
6E-0
4 -4
0 2.
39E-
05
35
2.9
-1.9
7E-0
4 5
-7.4
7E-0
9 -7
.47E
-05
-2.3
9E-0
3 -1
00
2.39
E-05
40
4.
02
2.21
E-03
5
8.36
E-08
8.
36E-
04
2.68
E-02
11
20
2.39
E-05
45
4.
31
5.71
E-04
5
2.17
E-08
2.
17E-
04
6.93
E-03
29
0 2.
39E-
05
50
3.8
-1.0
0E-0
3 5
-3.8
1E-0
8 -3
.81E
-04
-1.2
2E-0
2 -5
10
2.39
E-05
55
3.
79
-1.9
7E-0
5 5
-7.4
7E-1
0 -7
.47E
-06
-2.3
9E-0
4 -1
0 2.
39E-
05
60
3.69
-1
.97E
-04
5 -7
.47E
-09
-7.4
7E-0
5 -2
.39E
-03
-100
2.
39E-
05
65
3.63
-1
.18E
-04
5 -4
.48E
-09
-4.4
8E-0
5 -1
.43E
-03
-60
2.39
E-05
70
3.
44
-3.7
4E-0
4 5
-1.4
2E-0
8 -1
.42E
-04
-4.5
4E-0
3 -1
90
2.39
E-05
75
3.
4 -7
.88E
-05
5 -2
.99E
-09
-2.9
9E-0
5 -9
.56E
-04
-40
2.39
E-05
85
3.
51
2.17
E-04
10
4.
11E-
09
4.11
E-05
1.
31E-
03
110
1.19
E-05
95
3.
55
7.88
E-05
10
1.
49E-
09
1.49
E-05
4.
78E-
04
40
1.19
E-05
126
Tabl
e C
9. F
inal
DO
cal
cula
tions
. Sum
mar
y ta
ble
Flow
rate
A
ir Pr
essu
re
Max
imum
Flu
x ko
(ove
rall)
1/
ko (o
vera
ll)
1/kL
(liq
uid)
Sh
D
iffus
ion
[m
L/m
in]
[psi
] [m
mol
-DO
/m2 -m
in]
[m/m
in]
[min
/m]
[min
/m]
[-]
[m2 /m
in]
0.3
0.5
0.00
0278
821
7.96
631E
-06
1255
28.5
7 12
5528
.57
13
1.18
2E-0
7 1
0.5
0.00
0530
258
7.96
631E
-06
1255
28.5
7 12
5528
.57
13
1.18
2E-0
7 10
0.
5 0.
0001
9417
9 2.
3898
9E-0
5 41
842.
86
4184
2.86
40
1.
182E
-07
C
alcu
latin
g th
e am
ount
of o
xyge
n re
quire
d by
bio
film
at d
iffer
ent f
low
rate
s
NH
4 +1.
863O
2 + 0
.098
CO
2 → 0
.019
6 C
5H7N
O2 +
0.9
8 N
O3 +
0.0
941H
2O +
1.9
8 H
2O
Tota
l Mem
bran
e Su
rfac
e A
rea=
52
72.2
cm
2
O
xyge
n co
nsum
ed p
er a
mm
onia
use
d is
=
4.25
g
O2
Tabl
e C
10. B
iofil
m C
onsu
mpt
ion
Flow
rate
(m
L/m
in)
HR
T
(min
)
NH
4-N
co
nsum
ed
(g)
Bio
film
Oxy
gen
cons
umpt
ion
(m
mol
-DO
/sec
-m2 )
0.3
210
0.03
60
7.20
E-04
0.
5 12
6 0.
0549
1.
83E-
03
0.6
105
0.05
51
2.20
E-03
0.
7 90
0.
0621
2.
90E-
03
1 63
0.
0739
4.
93E-
03
1.5
42
0.11
19
1.12
E-02
1.
7 37
0.
0554
6.
27E-
03
127
APPENDIX D
TRACER STUDY CALCULATIONS
128
Tabl
e D
1. In
itial
trac
er a
t 0.3
mL/
min
time
t/T
(C/C
o)ex
p Pa
rt A
Pa
rt B
th
eore
tical
Co
sum
sq.e
rror
su
m
0 0
0.00
00
0 0
0 0
2.88
100
0.18
0.
0045
0.
252
1.14
4 0.
288
0.08
0
L =
0.2
m
120
0.21
0.
0057
0.
230
1.10
8 0.
255
0.06
2
u =
0.00
0009
m
/s
145
0.26
0.
0102
0.
209
1.07
8 0.
225
0.04
6
D =
0.
0000
1310
m
2 /sec
150
0.27
0.
0102
0.
206
1.07
4 0.
221
0.04
4
d =
7.03
165
0.29
0.
0159
0.
196
1.06
2 0.
208
0.03
7
Pe =
0.
14
175
0.31
0.
0193
0.
190
1.05
5 0.
201
0.03
3
180
0.32
0.
0375
0.
188
1.05
2 0.
197
0.02
6
20
0 0.
36
0.06
59
0.17
8 1.
042
0.18
5 0.
014
220
0.39
0.
1136
0.
170
1.03
4 0.
175
0.00
4
280
0.50
0.
6477
0.
150
1.01
8 0.
153
0.24
5
295
0.53
0.
7523
0.
147
1.01
5 0.
149
0.36
4
325
0.58
0.
7920
0.
140
1.01
1 0.
141
0.42
4
520
0.93
0.
9716
0.
110
1.00
0 0.
110
0.74
2
560
1 0.
9807
0.
106
1.00
0 0.
106
0.76
4
129
Tabl
e D
2. In
itial
trac
er a
t 1 m
L/m
in
time
t/T
(C/C
o)ex
p Pa
rt A
Pa
rt B
th
eore
tical
C
o su
m
sq.e
rror
su
m
0 0.
000
0.00
34
0.00
00
0.00
00
0.00
00
0.00
00
3.38
20
0.06
7 0.
0034
0.
3339
1.
3565
0.
4529
0.
2021
L =
0.2
m
35
0.11
7 0.
0091
0.
2524
1.
1689
0.
2950
0.
0818
u =
0.00
0031
m
/s
50
0.16
7 0.
0159
0.
2112
1.
1021
0.
2327
0.
0470
D =
0.
0000
7 m
2 /sec
60
0.20
0 0.
0466
0.
1928
1.
0775
0.
2077
0.
0260
d =
10.7
1
90
0.30
0 0.
7864
0.
1574
1.
0389
0.
1635
0.
3879
Pe =
0.
09
10
5 0.
350
0.89
77
0.14
57
1.02
86
0.14
99
0.55
93
120
0.40
0 0.
9284
0.
1363
1.
0212
0.
1392
0.
6228
125
0.41
7 0.
9375
0.
1336
1.
0192
0.
1361
0.
6422
300
1 0.
9864
0.
0862
1.
0000
0.
0862
0.
8103
130
Tabl
e D
3. In
itial
trac
er a
t 15
mL/
min
time
t/T
(C/C
o)ex
p Pa
rt A
Pa
rt B
th
eore
tical
C
o su
m
sq.e
rror
su
m
0 0.
000
0.00
0 0.
000
0.00
0 0.
000
0.00
0 3.
780
5 0.
053
0.55
9 0.
375
1.48
7 0.
558
0.00
0
10
0.10
5 0.
645
0.26
5 1.
194
0.31
7 0.
108
L
= 0.
2 m
15
0.15
8 0.
752
0.21
7 1.
110
0.24
1 0.
261
u
= 0.
0004
7 m
/s
20
0.21
1 0.
797
0.18
8 1.
071
0.20
1 0.
355
D
=
0.00
100
m2 /s
ec
25
0.26
3 0.
844
0.16
8 1.
049
0.17
6 0.
446
d
= 10
.74
35
0.36
8 0.
875
0.14
2 1.
026
0.14
5 0.
532
Pe
=
0.09
65
0.68
4 0.
914
0.10
4 1.
003
0.10
4 0.
655
80
0.84
2 0.
921
0.09
4 1.
001
0.09
4 0.
685
95
1.00
0 0.
945
0.08
6 1.
000
0.08
6 0.
738
131
Tabl
e D
4. F
inal
trac
er a
t 0.3
mL/
min
time
t/T
(C/C
o)ex
p Pa
rt A
Pa
rt B
th
eore
tical
Co
sum
sq.e
rror
su
m
0 0
0 0
0 0
0 4.
39
70
0.04
0.
272
0.43
3 1.
723
0.74
7 0.
226
105
0.06
0.
432
0.35
4 1.
416
0.50
1 0.
005
L
= 0.
2 m
140
0.08
0.
456
0.30
6 1.
285
0.39
4 0.
004
u
= 0.
0000
09
m/s
190
0.11
0.
650
0.26
3 1.
189
0.31
3 0.
114
D
=
0.00
0020
m
2 /sec
265
0.15
0.
696
0.22
3 1.
120
0.24
9 0.
199
d
= 10
.85
76
5 0.
43
0.91
9 0.
131
1.01
8 0.
133
0.61
8
Pe =
0.
09
835
0.47
0.
813
0.12
5 1.
014
0.12
7 0.
470
1540
0.
86
0.89
8 0.
092
1.00
1 0.
092
0.64
9
1600
0.
89
0.91
2 0.
091
1.00
0 0.
091
0.67
4
1780
0.
99
0.92
5 0.
086
1.00
0 0.
086
0.70
5
1790
1
0.93
6 0.
086
1.00
0 0.
086
0.72
3
132
Tabl
e D
5. F
inal
trac
er a
t 1 m
L/m
in
time
t/T
(C/C
o)ex
p Pa
rt A
Pa
rt B
th
eore
tical
Co
sum
sq.e
rror
su
m
0 0.
000
0 0
0 0
0 4.
1077
4201
6
15
0.04
5 0.
612
0.33
2 1.
372
0.45
6 0.
024
40
0.11
9 0.
714
0.20
3 1.
106
0.22
5 0.
239
L
= 0.
2 m
45
0.13
4 0.
780
0.19
2 1.
090
0.20
9 0.
326
u
= 0.
0000
31
m/s
75
0.22
4 0.
836
0.14
9 1.
043
0.15
5 0.
464
D
=
0.00
0100
m
2 /sec
11
0 0.
328
0.82
7 0.
123
1.02
2 0.
125
0.49
2
d =
16.1
1
140
0.41
8 0.
865
0.10
9 1.
013
0.11
0 0.
569
Pe
=
0.06
150
0.44
8 0.
903
0.10
5 1.
011
0.10
6 0.
635
200
0.59
7 0.
917
0.09
1 1.
004
0.09
1 0.
681
215
0.64
2 0.
910
0.08
8 1.
003
0.08
8 0.
675
335
1.00
0 0.
931
0.07
0 1.
000
0.07
0 0.
741
133
Tabl
e D
6. F
inal
trac
er a
t 15
mL/
min
time
t/T
(C/C
o)ex
p Pa
rt A
Pa
rt B
th
eore
tical
Co
sum
sq.e
rror
su
m
0
0.00
0 0.
000
0.00
0 0.
000
0.00
0 0.
000
0.26
7934
361
1
0.00
8 0.
174
0.67
9 4.
142
2.81
1 -2
.637
L =
0.2
m
2 0.
017
0.48
2 0.
480
2.01
1 0.
965
-0.4
83
u
= 0.
0004
66
m/s
3 0.
025
0.64
9 0.
392
1.58
0 0.
619
0.02
9
D =
0.
0019
0 m
2 /sec
4
0.03
4 0.
687
0.33
9 1.
401
0.47
5 0.
212
d
= 20
.41
8
0.06
8 0.
774
0.24
0 1.
170
0.28
1 0.
494
Pe
=
0.05
10
0.08
5 0.
814
0.21
5 1.
129
0.24
2 0.
572
13
0.11
0 0.
828
0.18
8 1.
092
0.20
6 0.
622
18
0.15
3 0.
868
0.16
0 1.
059
0.16
9 0.
699
23
0.19
5 0.
908
0.14
1 1.
042
0.14
7 0.
760
28
0.23
7 0.
828
0.12
8 1.
030
0.13
2 0.
696
33
0.28
0 0.
934
0.11
8 1.
023
0.12
1 0.
814
38
0.32
2 0.
946
0.11
0 1.
018
0.11
2 0.
834
53
0.44
9 0.
955
0.09
3 1.
008
0.09
4 0.
861
63
0.53
4 0.
946
0.08
5 1.
005
0.08
6 0.
860
73
0.61
9 0.
971
0.07
9 1.
003
0.08
0 0.
891
88
0.74
6 0.
992
0.07
2 1.
001
0.07
2 0.
920
108
0.91
5 0.
971
0.06
5 1.
000
0.06
5 0.
905
118
1.00
0 0.
980
0.06
2 1.
000
0.06
2 0.
917
134
135
APPENDIX E
T-TEST
Sam
ple
Cal
cula
tions
for T
-test
T-T
est
T*
=[x-
µ]/(S
/(n^0
.5))
W
here
t* is
the
devi
atio
n of
the
estim
ated
mea
n fr
om th
e po
pula
tion
mea
n, m
easu
red
in te
rms o
f the
un
it S/
(n^0
.5)
A lo
w v
alue
of t
* in
dica
tes l
ittle
diff
eren
ce b
etw
een
the
mea
ns
A
hig
h va
lue
of t*
indi
cate
s a la
rge
diff
eren
ce, p
rovi
ding
mor
e ju
stifi
catio
n fo
r rej
ectin
g th
e nu
ll hy
poth
esis
IF
t* >
tc, w
e re
ject
the
null
hypo
thes
is a
nd c
oncl
ude
ther
e is
a
diff
eren
ce
IF
t* <
tc, w
e ca
nnot
reje
ct th
e nu
ll hy
poth
esis
. The
evi
denc
e su
gges
ts th
e sa
mpl
e ha
s not
dev
iate
d fr
om th
e po
pula
tion
Fl
ow
Rat
e H
RT
(day
s)
%
Nitr
ifica
tion
0.3
0.15
74
.74
0.5
0.09
68
.87
0.6
0.07
51
.87
0.7
0.06
44
.17
1 0.
04
39.2
9
1.
5 0.
029
28.5
7
1.
7 0.
026
15.8
9
Mea
n =
46.2
Std.
Dev
iatio
n =
21.0
n =
7
(n-1
) =
6
tc (5
%) =
1.
943
(1 si
de
test
)
tc (1
%) =
3.
143
(1 si
de
test
)
136
13
8
For
the
95%
con
fiden
ce li
mits
t* (
74.7
) =
3.6
t* >
tc in
dica
tes s
tatis
tical
ly si
gnifi
cant
diff
eren
ce, n
ull h
ypot
hesi
s rej
ecte
d t*
( 68
.9) =
2.
9 t*
> tc
indi
cate
s sta
tistic
ally
sign
ifica
nt d
iffer
ence
, nul
l hyp
othe
sis r
ejec
ted
Fo
r th
e 99
% c
onfid
ence
lim
its
t*
( 74
.7) =
3.
6 t*
> tc
indi
cate
s sta
tistic
ally
sign
ifica
nt d
iffer
ence
, nul
l hyp
othe
sis r
ejec
ted
t* (
68.9
) =
2.9
t* <
tc te
st in
dica
tes i
nsuf
ficie
nt e
vide
nce
for a
stat
istic
ally
sign
ifica
nt
diff
eren
ce.
137
PERMISSION TO COPY
In presenting this thesis in partial fulfillment of the requirements for a master’s
degree at Texas Tech University or Texas Tech University Health Sciences Center, I
agree that the Library and my major department shall make it freely available for
research purposes. Permission to copy this thesis for scholarly purposes may be granted
by the Director of the Library or my major professor. It is understood that any copying
or publication of this thesis for financial gain shall not be allowed without my further
written permission and that any user may be liable for copyright infringement.
Agree (Permission is granted.)
Maria Noel Ruiz Careri 03/08/2005 Student Signature Date Disagree (Permission is not granted.) _______________________________________________ ____________ Student Signature Date